Resguardo inicial

linux-7.2.3/drivers/iommu/iommufd/Kconfig

@@ -0,0 +1,51 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config IOMMUFD_DRIVER_CORE
+	bool
+	default (IOMMUFD_DRIVER || IOMMUFD) if IOMMUFD!=n
+
+config IOMMUFD
+	tristate "IOMMU Userspace API"
+	select INTERVAL_TREE
+	select INTERVAL_TREE_SPAN_ITER
+	select IOMMU_API
+	default n
+	help
+	  Provides /dev/iommu, the user API to control the IOMMU subsystem as
+	  it relates to managing IO page tables that point at user space memory.
+
+	  If you don't know what to do here, say N.
+
+if IOMMUFD
+config IOMMUFD_VFIO_CONTAINER
+	bool "IOMMUFD provides the VFIO container /dev/vfio/vfio"
+	depends on VFIO_GROUP && !VFIO_CONTAINER
+	default VFIO_GROUP && !VFIO_CONTAINER
+	help
+	  IOMMUFD will provide /dev/vfio/vfio instead of VFIO. This relies on
+	  IOMMUFD providing compatibility emulation to give the same ioctls.
+	  It provides an option to build a kernel with legacy VFIO components
+	  removed.
+
+	  IOMMUFD VFIO container emulation is known to lack certain features
+	  of the native VFIO container, such as peer-to-peer
+	  DMA mapping, PPC IOMMU support, as well as other potentially
+	  undiscovered gaps.  This option is currently intended for the
+	  purpose of testing IOMMUFD with unmodified userspace supporting VFIO
+	  and making use of the Type1 VFIO IOMMU backend.  General purpose
+	  enabling of this option is currently discouraged.
+
+	  Unless testing IOMMUFD, say N here.
+
+config IOMMUFD_TEST
+	bool "IOMMU Userspace API Test support"
+	depends on DEBUG_KERNEL
+	depends on FAULT_INJECTION
+	depends on RUNTIME_TESTING_MENU
+	depends on IOMMU_PT_AMDV1=y || IOMMUFD=IOMMU_PT_AMDV1
+	select DMA_SHARED_BUFFER
+	select IOMMUFD_DRIVER
+	default n
+	help
+	  This is dangerous, do not enable unless running
+	  tools/testing/selftests/iommu
+endif

linux-7.2.3/drivers/iommu/iommufd/Makefile

@@ -0,0 +1,19 @@
+# SPDX-License-Identifier: GPL-2.0-only
+iommufd-y := \
+	device.o \
+	eventq.o \
+	hw_pagetable.o \
+	io_pagetable.o \
+	ioas.o \
+	main.o \
+	pages.o \
+	vfio_compat.o \
+	viommu.o
+
+iommufd-$(CONFIG_IOMMUFD_TEST) += selftest.o
+
+obj-$(CONFIG_IOMMUFD) += iommufd.o
+obj-$(CONFIG_IOMMUFD_DRIVER) += iova_bitmap.o
+
+iommufd_driver-y := driver.o
+obj-$(CONFIG_IOMMUFD_DRIVER_CORE) += iommufd_driver.o

linux-7.2.3/drivers/iommu/iommufd/double_span.h

@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES.
+ */
+#ifndef __IOMMUFD_DOUBLE_SPAN_H
+#define __IOMMUFD_DOUBLE_SPAN_H
+
+#include <linux/interval_tree.h>
+
+/*
+ * This is a variation of the general interval_tree_span_iter that computes the
+ * spans over the union of two different interval trees. Used ranges are broken
+ * up and reported based on the tree that provides the interval. The first span
+ * always takes priority. Like interval_tree_span_iter it is greedy and the same
+ * value of is_used will not repeat on two iteration cycles.
+ */
+struct interval_tree_double_span_iter {
+	struct rb_root_cached *itrees[2];
+	struct interval_tree_span_iter spans[2];
+	union {
+		unsigned long start_hole;
+		unsigned long start_used;
+	};
+	union {
+		unsigned long last_hole;
+		unsigned long last_used;
+	};
+	/* 0 = hole, 1 = used span[0], 2 = used span[1], -1 done iteration */
+	int is_used;
+};
+
+void interval_tree_double_span_iter_update(
+	struct interval_tree_double_span_iter *iter);
+void interval_tree_double_span_iter_first(
+	struct interval_tree_double_span_iter *iter,
+	struct rb_root_cached *itree1, struct rb_root_cached *itree2,
+	unsigned long first_index, unsigned long last_index);
+void interval_tree_double_span_iter_next(
+	struct interval_tree_double_span_iter *iter);
+
+static inline bool
+interval_tree_double_span_iter_done(struct interval_tree_double_span_iter *state)
+{
+	return state->is_used == -1;
+}
+
+#define interval_tree_for_each_double_span(span, itree1, itree2, first_index, \
+					   last_index)                        \
+	for (interval_tree_double_span_iter_first(span, itree1, itree2,       \
+						  first_index, last_index);   \
+	     !interval_tree_double_span_iter_done(span);                      \
+	     interval_tree_double_span_iter_next(span))
+
+#endif

linux-7.2.3/drivers/iommu/iommufd/driver.c

@@ -0,0 +1,307 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2024, NVIDIA CORPORATION & AFFILIATES
+ */
+#include "iommufd_private.h"
+
+/* Driver should use a per-structure helper in include/linux/iommufd.h */
+int _iommufd_object_depend(struct iommufd_object *obj_dependent,
+			   struct iommufd_object *obj_depended)
+{
+	/* Reject self dependency that dead locks */
+	if (obj_dependent == obj_depended)
+		return -EINVAL;
+	/* Only support dependency between two objects of the same type */
+	if (obj_dependent->type != obj_depended->type)
+		return -EINVAL;
+
+	refcount_inc(&obj_depended->users);
+	return 0;
+}
+EXPORT_SYMBOL_NS_GPL(_iommufd_object_depend, "IOMMUFD");
+
+/* Driver should use a per-structure helper in include/linux/iommufd.h */
+void _iommufd_object_undepend(struct iommufd_object *obj_dependent,
+			      struct iommufd_object *obj_depended)
+{
+	if (WARN_ON_ONCE(obj_dependent == obj_depended ||
+			 obj_dependent->type != obj_depended->type))
+		return;
+
+	refcount_dec(&obj_depended->users);
+}
+EXPORT_SYMBOL_NS_GPL(_iommufd_object_undepend, "IOMMUFD");
+
+/*
+ * Allocate an @offset to return to user space to use for an mmap() syscall
+ *
+ * Driver should use a per-structure helper in include/linux/iommufd.h
+ */
+int _iommufd_alloc_mmap(struct iommufd_ctx *ictx, struct iommufd_object *owner,
+			phys_addr_t mmio_addr, size_t length,
+			unsigned long *offset)
+{
+	struct iommufd_mmap *immap;
+	unsigned long startp;
+	int rc;
+
+	if (!PAGE_ALIGNED(mmio_addr))
+		return -EINVAL;
+	if (!length || !PAGE_ALIGNED(length))
+		return -EINVAL;
+
+	immap = kzalloc(sizeof(*immap), GFP_KERNEL);
+	if (!immap)
+		return -ENOMEM;
+	immap->owner = owner;
+	immap->length = length;
+	immap->mmio_addr = mmio_addr;
+
+	/* Skip the first page to ease caller identifying the returned offset */
+	rc = mtree_alloc_range(&ictx->mt_mmap, &startp, immap, immap->length,
+			       PAGE_SIZE, ULONG_MAX, GFP_KERNEL);
+	if (rc < 0) {
+		kfree(immap);
+		return rc;
+	}
+
+	/* mmap() syscall will right-shift the offset in vma->vm_pgoff too */
+	immap->vm_pgoff = startp >> PAGE_SHIFT;
+	*offset = startp;
+	return 0;
+}
+EXPORT_SYMBOL_NS_GPL(_iommufd_alloc_mmap, "IOMMUFD");
+
+/* Driver should use a per-structure helper in include/linux/iommufd.h */
+void _iommufd_destroy_mmap(struct iommufd_ctx *ictx,
+			   struct iommufd_object *owner, unsigned long offset)
+{
+	struct iommufd_mmap *immap;
+
+	immap = mtree_erase(&ictx->mt_mmap, offset);
+	WARN_ON_ONCE(!immap || immap->owner != owner);
+	kfree(immap);
+}
+EXPORT_SYMBOL_NS_GPL(_iommufd_destroy_mmap, "IOMMUFD");
+
+struct device *iommufd_vdevice_to_device(struct iommufd_vdevice *vdev)
+{
+	return vdev->idev->dev;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_vdevice_to_device, "IOMMUFD");
+
+/* Caller should xa_lock(&viommu->vdevs) to protect the return value */
+struct device *iommufd_viommu_find_dev(struct iommufd_viommu *viommu,
+				       unsigned long vdev_id)
+{
+	struct iommufd_vdevice *vdev;
+
+	lockdep_assert_held(&viommu->vdevs.xa_lock);
+
+	vdev = xa_load(&viommu->vdevs, vdev_id);
+	return vdev ? iommufd_vdevice_to_device(vdev) : NULL;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_viommu_find_dev, "IOMMUFD");
+
+/* Return -ENOENT if device is not associated to the vIOMMU */
+int iommufd_viommu_get_vdev_id(struct iommufd_viommu *viommu,
+			       struct device *dev, unsigned long *vdev_id)
+{
+	struct iommufd_vdevice *vdev;
+	unsigned long index;
+	int rc = -ENOENT;
+
+	if (WARN_ON_ONCE(!vdev_id))
+		return -EINVAL;
+
+	xa_lock(&viommu->vdevs);
+	xa_for_each(&viommu->vdevs, index, vdev) {
+		if (iommufd_vdevice_to_device(vdev) == dev) {
+			*vdev_id = vdev->virt_id;
+			rc = 0;
+			break;
+		}
+	}
+	xa_unlock(&viommu->vdevs);
+	return rc;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_viommu_get_vdev_id, "IOMMUFD");
+
+/*
+ * Typically called in driver's threaded IRQ handler.
+ * The @type and @event_data must be defined in include/uapi/linux/iommufd.h
+ */
+int iommufd_viommu_report_event(struct iommufd_viommu *viommu,
+				enum iommu_veventq_type type, void *event_data,
+				size_t data_len)
+{
+	struct iommufd_veventq *veventq;
+	struct iommufd_vevent *vevent;
+	int rc = 0;
+
+	if (WARN_ON_ONCE(!data_len || !event_data))
+		return -EINVAL;
+
+	down_read(&viommu->veventqs_rwsem);
+
+	veventq = iommufd_viommu_find_veventq(viommu, type);
+	if (!veventq) {
+		rc = -EOPNOTSUPP;
+		goto out_unlock_veventqs;
+	}
+
+	/* Pre-allocate to avoid GFP_ATOMIC; use GFP_NOWAIT to avoid sleeping */
+	vevent = kzalloc_flex(*vevent, event_data, data_len, GFP_NOWAIT);
+	if (!vevent) {
+		spin_lock(&veventq->common.lock);
+		vevent = &veventq->lost_events_header;
+		rc = -ENOMEM;
+		goto out_set_header;
+	}
+
+	spin_lock(&veventq->common.lock);
+	if (veventq->num_events == veventq->depth) {
+		kfree(vevent);
+		vevent = &veventq->lost_events_header;
+		goto out_set_header;
+	}
+	vevent->data_len = data_len;
+	memcpy(vevent->event_data, event_data, data_len);
+	veventq->num_events++;
+
+out_set_header:
+	iommufd_vevent_handler(veventq, vevent);
+	spin_unlock(&veventq->common.lock);
+out_unlock_veventqs:
+	up_read(&viommu->veventqs_rwsem);
+	return rc;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_viommu_report_event, "IOMMUFD");
+
+#ifdef CONFIG_IRQ_MSI_IOMMU
+/*
+ * Get a iommufd_sw_msi_map for the msi physical address requested by the irq
+ * layer. The mapping to IOVA is global to the iommufd file descriptor, every
+ * domain that is attached to a device using the same MSI parameters will use
+ * the same IOVA.
+ */
+static struct iommufd_sw_msi_map *
+iommufd_sw_msi_get_map(struct iommufd_ctx *ictx, phys_addr_t msi_addr,
+		       phys_addr_t sw_msi_start)
+{
+	struct iommufd_sw_msi_map *cur;
+	unsigned int max_pgoff = 0;
+
+	lockdep_assert_held(&ictx->sw_msi_lock);
+
+	list_for_each_entry(cur, &ictx->sw_msi_list, sw_msi_item) {
+		if (cur->sw_msi_start != sw_msi_start)
+			continue;
+		max_pgoff = max(max_pgoff, cur->pgoff + 1);
+		if (cur->msi_addr == msi_addr)
+			return cur;
+	}
+
+	if (ictx->sw_msi_id >=
+	    BITS_PER_BYTE * sizeof_field(struct iommufd_sw_msi_maps, bitmap))
+		return ERR_PTR(-EOVERFLOW);
+
+	cur = kzalloc_obj(*cur);
+	if (!cur)
+		return ERR_PTR(-ENOMEM);
+
+	cur->sw_msi_start = sw_msi_start;
+	cur->msi_addr = msi_addr;
+	cur->pgoff = max_pgoff;
+	cur->id = ictx->sw_msi_id++;
+	list_add_tail(&cur->sw_msi_item, &ictx->sw_msi_list);
+	return cur;
+}
+
+int iommufd_sw_msi_install(struct iommufd_ctx *ictx,
+			   struct iommufd_hwpt_paging *hwpt_paging,
+			   struct iommufd_sw_msi_map *msi_map)
+{
+	unsigned long iova;
+
+	lockdep_assert_held(&ictx->sw_msi_lock);
+
+	iova = msi_map->sw_msi_start + msi_map->pgoff * PAGE_SIZE;
+	if (!test_bit(msi_map->id, hwpt_paging->present_sw_msi.bitmap)) {
+		int rc;
+
+		rc = iommu_map(hwpt_paging->common.domain, iova,
+			       msi_map->msi_addr, PAGE_SIZE,
+			       IOMMU_WRITE | IOMMU_READ | IOMMU_MMIO,
+			       GFP_KERNEL_ACCOUNT);
+		if (rc)
+			return rc;
+		__set_bit(msi_map->id, hwpt_paging->present_sw_msi.bitmap);
+	}
+	return 0;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_sw_msi_install, "IOMMUFD_INTERNAL");
+
+/*
+ * Called by the irq code if the platform translates the MSI address through the
+ * IOMMU. msi_addr is the physical address of the MSI page. iommufd will
+ * allocate a fd global iova for the physical page that is the same on all
+ * domains and devices.
+ */
+int iommufd_sw_msi(struct iommu_domain *domain, struct msi_desc *desc,
+		   phys_addr_t msi_addr)
+{
+	struct device *dev = msi_desc_to_dev(desc);
+	struct iommufd_hwpt_paging *hwpt_paging;
+	struct iommu_attach_handle *raw_handle;
+	struct iommufd_attach_handle *handle;
+	struct iommufd_sw_msi_map *msi_map;
+	struct iommufd_ctx *ictx;
+	unsigned long iova;
+	int rc;
+
+	/*
+	 * It is safe to call iommu_attach_handle_get() here because the iommu
+	 * core code invokes this under the group mutex which also prevents any
+	 * change of the attach handle for the duration of this function.
+	 */
+	iommu_group_mutex_assert(dev);
+
+	raw_handle =
+		iommu_attach_handle_get(dev->iommu_group, IOMMU_NO_PASID, 0);
+	if (IS_ERR(raw_handle))
+		return 0;
+	hwpt_paging = find_hwpt_paging(domain->iommufd_hwpt);
+
+	handle = to_iommufd_handle(raw_handle);
+	/* No IOMMU_RESV_SW_MSI means no change to the msi_msg */
+	if (handle->idev->igroup->sw_msi_start == PHYS_ADDR_MAX)
+		return 0;
+
+	ictx = handle->idev->ictx;
+	guard(mutex)(&ictx->sw_msi_lock);
+	/*
+	 * The input msi_addr is the exact byte offset of the MSI doorbell, we
+	 * assume the caller has checked that it is contained with a MMIO region
+	 * that is secure to map at PAGE_SIZE.
+	 */
+	msi_map = iommufd_sw_msi_get_map(handle->idev->ictx,
+					 msi_addr & PAGE_MASK,
+					 handle->idev->igroup->sw_msi_start);
+	if (IS_ERR(msi_map))
+		return PTR_ERR(msi_map);
+
+	rc = iommufd_sw_msi_install(ictx, hwpt_paging, msi_map);
+	if (rc)
+		return rc;
+	__set_bit(msi_map->id, handle->idev->igroup->required_sw_msi.bitmap);
+
+	iova = msi_map->sw_msi_start + msi_map->pgoff * PAGE_SIZE;
+	msi_desc_set_iommu_msi_iova(desc, iova, PAGE_SHIFT);
+	return 0;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_sw_msi, "IOMMUFD");
+#endif
+
+MODULE_DESCRIPTION("iommufd code shared with builtin modules");
+MODULE_IMPORT_NS("IOMMUFD_INTERNAL");
+MODULE_LICENSE("GPL");

linux-7.2.3/drivers/iommu/iommufd/eventq.c

@@ -0,0 +1,570 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2024 Intel Corporation
+ */
+#define pr_fmt(fmt) "iommufd: " fmt
+
+#include <linux/anon_inodes.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/iommufd.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/poll.h>
+#include <uapi/linux/iommufd.h>
+
+#include "../iommu-priv.h"
+#include "iommufd_private.h"
+
+/* IOMMUFD_OBJ_FAULT Functions */
+void iommufd_auto_response_faults(struct iommufd_hw_pagetable *hwpt,
+				  struct iommufd_attach_handle *handle)
+{
+	struct iommufd_fault *fault = hwpt->fault;
+	struct iopf_group *group, *next;
+	struct list_head free_list;
+	unsigned long index;
+
+	if (!fault || !handle)
+		return;
+	INIT_LIST_HEAD(&free_list);
+
+	mutex_lock(&fault->mutex);
+	spin_lock(&fault->common.lock);
+	list_for_each_entry_safe(group, next, &fault->common.deliver, node) {
+		if (group->attach_handle != &handle->handle)
+			continue;
+		list_move(&group->node, &free_list);
+	}
+	spin_unlock(&fault->common.lock);
+
+	list_for_each_entry_safe(group, next, &free_list, node) {
+		list_del(&group->node);
+		iopf_group_response(group, IOMMU_PAGE_RESP_INVALID);
+		iopf_free_group(group);
+	}
+
+	xa_for_each(&fault->response, index, group) {
+		if (group->attach_handle != &handle->handle)
+			continue;
+		xa_erase(&fault->response, index);
+		iopf_group_response(group, IOMMU_PAGE_RESP_INVALID);
+		iopf_free_group(group);
+	}
+	mutex_unlock(&fault->mutex);
+}
+
+void iommufd_fault_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_eventq *eventq =
+		container_of(obj, struct iommufd_eventq, obj);
+	struct iommufd_fault *fault = eventq_to_fault(eventq);
+	struct iopf_group *group, *next;
+	unsigned long index;
+
+	/*
+	 * The iommufd object's reference count is zero at this point.
+	 * We can be confident that no other threads are currently
+	 * accessing this pointer. Therefore, acquiring the mutex here
+	 * is unnecessary.
+	 */
+	list_for_each_entry_safe(group, next, &fault->common.deliver, node) {
+		list_del(&group->node);
+		iopf_group_response(group, IOMMU_PAGE_RESP_INVALID);
+		iopf_free_group(group);
+	}
+	xa_for_each(&fault->response, index, group) {
+		xa_erase(&fault->response, index);
+		iopf_group_response(group, IOMMU_PAGE_RESP_INVALID);
+		iopf_free_group(group);
+	}
+	xa_destroy(&fault->response);
+	mutex_destroy(&fault->mutex);
+}
+
+static void iommufd_compose_fault_message(struct iommu_fault *fault,
+					  struct iommu_hwpt_pgfault *hwpt_fault,
+					  struct iommufd_device *idev,
+					  u32 cookie)
+{
+	hwpt_fault->flags = fault->prm.flags;
+	hwpt_fault->dev_id = idev->obj.id;
+	hwpt_fault->pasid = fault->prm.pasid;
+	hwpt_fault->grpid = fault->prm.grpid;
+	hwpt_fault->perm = fault->prm.perm;
+	hwpt_fault->addr = fault->prm.addr;
+	hwpt_fault->length = 0;
+	hwpt_fault->cookie = cookie;
+}
+
+/* Fetch the first node out of the fault->deliver list */
+static struct iopf_group *
+iommufd_fault_deliver_fetch(struct iommufd_fault *fault)
+{
+	struct list_head *list = &fault->common.deliver;
+	struct iopf_group *group = NULL;
+
+	spin_lock(&fault->common.lock);
+	if (!list_empty(list)) {
+		group = list_first_entry(list, struct iopf_group, node);
+		list_del(&group->node);
+	}
+	spin_unlock(&fault->common.lock);
+	return group;
+}
+
+/* Restore a node back to the head of the fault->deliver list */
+static void iommufd_fault_deliver_restore(struct iommufd_fault *fault,
+					  struct iopf_group *group)
+{
+	spin_lock(&fault->common.lock);
+	list_add(&group->node, &fault->common.deliver);
+	spin_unlock(&fault->common.lock);
+}
+
+static ssize_t iommufd_fault_fops_read(struct file *filep, char __user *buf,
+				       size_t count, loff_t *ppos)
+{
+	size_t fault_size = sizeof(struct iommu_hwpt_pgfault);
+	struct iommufd_eventq *eventq = filep->private_data;
+	struct iommufd_fault *fault = eventq_to_fault(eventq);
+	struct iommu_hwpt_pgfault data = {};
+	struct iommufd_device *idev;
+	struct iopf_group *group;
+	struct iopf_fault *iopf;
+	size_t done = 0;
+	int rc = 0;
+
+	if (*ppos || count % fault_size)
+		return -ESPIPE;
+
+	mutex_lock(&fault->mutex);
+	while ((group = iommufd_fault_deliver_fetch(fault))) {
+		size_t group_done = done;
+
+		if (done >= count ||
+		    group->fault_count * fault_size > count - done) {
+			iommufd_fault_deliver_restore(fault, group);
+			/* Read count doesn't fit the first fault group */
+			if (done == 0)
+				rc = -EINVAL;
+			break;
+		}
+
+		rc = xa_alloc(&fault->response, &group->cookie, group,
+			      xa_limit_32b, GFP_KERNEL);
+		if (rc) {
+			iommufd_fault_deliver_restore(fault, group);
+			break;
+		}
+
+		idev = to_iommufd_handle(group->attach_handle)->idev;
+		list_for_each_entry(iopf, &group->faults, list) {
+			iommufd_compose_fault_message(&iopf->fault,
+						      &data, idev,
+						      group->cookie);
+			if (copy_to_user(buf + group_done, &data, fault_size)) {
+				xa_erase(&fault->response, group->cookie);
+				iommufd_fault_deliver_restore(fault, group);
+				rc = -EFAULT;
+				break;
+			}
+			group_done += fault_size;
+		}
+		if (rc)
+			break;
+		done = group_done;
+	}
+	mutex_unlock(&fault->mutex);
+
+	return done == 0 ? rc : done;
+}
+
+static ssize_t iommufd_fault_fops_write(struct file *filep, const char __user *buf,
+					size_t count, loff_t *ppos)
+{
+	size_t response_size = sizeof(struct iommu_hwpt_page_response);
+	struct iommufd_eventq *eventq = filep->private_data;
+	struct iommufd_fault *fault = eventq_to_fault(eventq);
+	struct iommu_hwpt_page_response response;
+	struct iopf_group *group;
+	size_t done = 0;
+	int rc = 0;
+
+	if (*ppos || count % response_size)
+		return -ESPIPE;
+
+	mutex_lock(&fault->mutex);
+	while (count > done) {
+		if (copy_from_user(&response, buf + done, response_size)) {
+			rc = -EFAULT;
+			break;
+		}
+
+		static_assert((int)IOMMUFD_PAGE_RESP_SUCCESS ==
+			      (int)IOMMU_PAGE_RESP_SUCCESS);
+		static_assert((int)IOMMUFD_PAGE_RESP_INVALID ==
+			      (int)IOMMU_PAGE_RESP_INVALID);
+		if (response.code != IOMMUFD_PAGE_RESP_SUCCESS &&
+		    response.code != IOMMUFD_PAGE_RESP_INVALID) {
+			rc = -EINVAL;
+			break;
+		}
+
+		group = xa_erase(&fault->response, response.cookie);
+		if (!group) {
+			rc = -EINVAL;
+			break;
+		}
+
+		iopf_group_response(group, response.code);
+		iopf_free_group(group);
+		done += response_size;
+	}
+	mutex_unlock(&fault->mutex);
+
+	return done == 0 ? rc : done;
+}
+
+/* IOMMUFD_OBJ_VEVENTQ Functions */
+
+void iommufd_veventq_abort(struct iommufd_object *obj)
+{
+	struct iommufd_eventq *eventq =
+		container_of(obj, struct iommufd_eventq, obj);
+	struct iommufd_veventq *veventq = eventq_to_veventq(eventq);
+	struct iommufd_viommu *viommu = veventq->viommu;
+	struct iommufd_vevent *cur, *next;
+
+	lockdep_assert_held_write(&viommu->veventqs_rwsem);
+
+	list_for_each_entry_safe(cur, next, &eventq->deliver, node) {
+		list_del(&cur->node);
+		if (cur != &veventq->lost_events_header)
+			kfree(cur);
+	}
+
+	refcount_dec(&viommu->obj.users);
+	list_del(&veventq->node);
+}
+
+void iommufd_veventq_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_veventq *veventq = eventq_to_veventq(
+		container_of(obj, struct iommufd_eventq, obj));
+
+	down_write(&veventq->viommu->veventqs_rwsem);
+	iommufd_veventq_abort(obj);
+	up_write(&veventq->viommu->veventqs_rwsem);
+}
+
+static struct iommufd_vevent *
+iommufd_veventq_deliver_fetch(struct iommufd_veventq *veventq)
+{
+	struct iommufd_eventq *eventq = &veventq->common;
+	struct list_head *list = &eventq->deliver;
+	struct iommufd_vevent *vevent = NULL;
+
+	spin_lock(&eventq->lock);
+	if (!list_empty(list)) {
+		struct iommufd_vevent *next;
+
+		next = list_first_entry(list, struct iommufd_vevent, node);
+		/* Make a copy of the lost_events_header for copy_to_user */
+		if (next == &veventq->lost_events_header) {
+			vevent = kzalloc_obj(*vevent, GFP_ATOMIC);
+			if (!vevent) {
+				vevent = ERR_PTR(-ENOMEM);
+				goto out_unlock;
+			}
+		}
+		list_del(&next->node);
+		if (vevent)
+			memcpy(vevent, next, sizeof(*vevent));
+		else
+			vevent = next;
+	}
+out_unlock:
+	spin_unlock(&eventq->lock);
+	return vevent;
+}
+
+static void iommufd_veventq_deliver_restore(struct iommufd_veventq *veventq,
+					    struct iommufd_vevent *vevent)
+{
+	struct iommufd_eventq *eventq = &veventq->common;
+	struct list_head *list = &eventq->deliver;
+
+	spin_lock(&eventq->lock);
+	if (vevent_for_lost_events_header(vevent)) {
+		/* Remove the copy of the lost_events_header */
+		kfree(vevent);
+		vevent = NULL;
+		/* An empty list needs the lost_events_header back */
+		if (list_empty(list))
+			vevent = &veventq->lost_events_header;
+	}
+	if (vevent)
+		list_add(&vevent->node, list);
+	spin_unlock(&eventq->lock);
+}
+
+static ssize_t iommufd_veventq_fops_read(struct file *filep, char __user *buf,
+					 size_t count, loff_t *ppos)
+{
+	struct iommufd_eventq *eventq = filep->private_data;
+	struct iommufd_veventq *veventq = eventq_to_veventq(eventq);
+	struct iommufd_vevent_header *hdr;
+	struct iommufd_vevent *cur;
+	size_t done = 0;
+	int rc = 0;
+
+	if (*ppos)
+		return -ESPIPE;
+	/* Minimum read count is a vEVENT header */
+	if (count < sizeof(*hdr))
+		return -EINVAL;
+
+	while ((cur = iommufd_veventq_deliver_fetch(veventq))) {
+		if (IS_ERR(cur)) {
+			if (done == 0)
+				rc = PTR_ERR(cur);
+			break;
+		}
+
+		/* Validate the remaining bytes against the header size */
+		if (done >= count || sizeof(*hdr) > count - done) {
+			iommufd_veventq_deliver_restore(veventq, cur);
+			break;
+		}
+		hdr = &cur->header;
+
+		/* If being a normal vEVENT, validate against the full size */
+		if (!vevent_for_lost_events_header(cur) &&
+		    sizeof(*hdr) + cur->data_len > count - done) {
+			iommufd_veventq_deliver_restore(veventq, cur);
+			/* Read count doesn't fit a single normal vEVENT */
+			if (done == 0)
+				rc = -EINVAL;
+			break;
+		}
+
+		if (copy_to_user(buf + done, hdr, sizeof(*hdr))) {
+			iommufd_veventq_deliver_restore(veventq, cur);
+			rc = -EFAULT;
+			break;
+		}
+		done += sizeof(*hdr);
+
+		if (cur->data_len &&
+		    copy_to_user(buf + done, cur->event_data, cur->data_len)) {
+			iommufd_veventq_deliver_restore(veventq, cur);
+			done -= sizeof(*hdr);
+			rc = -EFAULT;
+			break;
+		}
+		spin_lock(&eventq->lock);
+		if (!vevent_for_lost_events_header(cur))
+			veventq->num_events--;
+		spin_unlock(&eventq->lock);
+		done += cur->data_len;
+		kfree(cur);
+	}
+
+	return done == 0 ? rc : done;
+}
+
+/* Common Event Queue Functions */
+
+static __poll_t iommufd_eventq_fops_poll(struct file *filep,
+					 struct poll_table_struct *wait)
+{
+	struct iommufd_eventq *eventq = filep->private_data;
+	__poll_t pollflags = 0;
+
+	if (eventq->obj.type == IOMMUFD_OBJ_FAULT)
+		pollflags |= EPOLLOUT;
+
+	poll_wait(filep, &eventq->wait_queue, wait);
+	spin_lock(&eventq->lock);
+	if (!list_empty(&eventq->deliver))
+		pollflags |= EPOLLIN | EPOLLRDNORM;
+	spin_unlock(&eventq->lock);
+
+	return pollflags;
+}
+
+static int iommufd_eventq_fops_release(struct inode *inode, struct file *filep)
+{
+	struct iommufd_eventq *eventq = filep->private_data;
+
+	refcount_dec(&eventq->obj.users);
+	iommufd_ctx_put(eventq->ictx);
+	return 0;
+}
+
+#define INIT_EVENTQ_FOPS(read_op, write_op)                                    \
+	((const struct file_operations){                                       \
+		.owner = THIS_MODULE,                                          \
+		.open = nonseekable_open,                                      \
+		.read = read_op,                                               \
+		.write = write_op,                                             \
+		.poll = iommufd_eventq_fops_poll,                              \
+		.release = iommufd_eventq_fops_release,                        \
+	})
+
+static int iommufd_eventq_init(struct iommufd_eventq *eventq, char *name,
+			       struct iommufd_ctx *ictx,
+			       const struct file_operations *fops)
+{
+	struct file *filep;
+
+	spin_lock_init(&eventq->lock);
+	INIT_LIST_HEAD(&eventq->deliver);
+	init_waitqueue_head(&eventq->wait_queue);
+
+	/* The filep is fput() by the core code during failure */
+	filep = anon_inode_getfile(name, fops, eventq, O_RDWR);
+	if (IS_ERR(filep))
+		return PTR_ERR(filep);
+
+	eventq->ictx = ictx;
+	iommufd_ctx_get(eventq->ictx);
+	eventq->filep = filep;
+	refcount_inc(&eventq->obj.users);
+
+	return get_unused_fd_flags(O_CLOEXEC);
+}
+
+static const struct file_operations iommufd_fault_fops =
+	INIT_EVENTQ_FOPS(iommufd_fault_fops_read, iommufd_fault_fops_write);
+
+int iommufd_fault_alloc(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_fault_alloc *cmd = ucmd->cmd;
+	struct iommufd_fault *fault;
+	int fdno;
+	int rc;
+
+	if (cmd->flags)
+		return -EOPNOTSUPP;
+
+	fault = __iommufd_object_alloc_ucmd(ucmd, fault, IOMMUFD_OBJ_FAULT,
+					    common.obj);
+	if (IS_ERR(fault))
+		return PTR_ERR(fault);
+
+	xa_init_flags(&fault->response, XA_FLAGS_ALLOC1);
+	mutex_init(&fault->mutex);
+
+	fdno = iommufd_eventq_init(&fault->common, "[iommufd-pgfault]",
+				   ucmd->ictx, &iommufd_fault_fops);
+	if (fdno < 0)
+		return fdno;
+
+	cmd->out_fault_id = fault->common.obj.id;
+	cmd->out_fault_fd = fdno;
+
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+	if (rc)
+		goto out_put_fdno;
+
+	fd_install(fdno, fault->common.filep);
+
+	return 0;
+out_put_fdno:
+	put_unused_fd(fdno);
+	return rc;
+}
+
+int iommufd_fault_iopf_handler(struct iopf_group *group)
+{
+	struct iommufd_hw_pagetable *hwpt;
+	struct iommufd_fault *fault;
+
+	hwpt = group->attach_handle->domain->iommufd_hwpt;
+	fault = hwpt->fault;
+
+	iopf_group_dequeue(group);
+
+	spin_lock(&fault->common.lock);
+	list_add_tail(&group->node, &fault->common.deliver);
+	spin_unlock(&fault->common.lock);
+
+	wake_up_interruptible(&fault->common.wait_queue);
+
+	return 0;
+}
+
+static const struct file_operations iommufd_veventq_fops =
+	INIT_EVENTQ_FOPS(iommufd_veventq_fops_read, NULL);
+
+/* An arbitrary upper bound for veventq_depth that fits all existing HWs */
+#define VEVENTQ_MAX_DEPTH (1U << 19)
+
+int iommufd_veventq_alloc(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_veventq_alloc *cmd = ucmd->cmd;
+	struct iommufd_veventq *veventq;
+	struct iommufd_viommu *viommu;
+	int fdno;
+	int rc;
+
+	if (cmd->flags || cmd->__reserved ||
+	    cmd->type == IOMMU_VEVENTQ_TYPE_DEFAULT)
+		return -EOPNOTSUPP;
+	if (!cmd->veventq_depth || cmd->veventq_depth > VEVENTQ_MAX_DEPTH)
+		return -EINVAL;
+
+	viommu = iommufd_get_viommu(ucmd, cmd->viommu_id);
+	if (IS_ERR(viommu))
+		return PTR_ERR(viommu);
+
+	down_write(&viommu->veventqs_rwsem);
+
+	if (iommufd_viommu_find_veventq(viommu, cmd->type)) {
+		rc = -EEXIST;
+		goto out_unlock_veventqs;
+	}
+
+	veventq = __iommufd_object_alloc(ucmd->ictx, veventq,
+					 IOMMUFD_OBJ_VEVENTQ, common.obj);
+	if (IS_ERR(veventq)) {
+		rc = PTR_ERR(veventq);
+		goto out_unlock_veventqs;
+	}
+
+	veventq->type = cmd->type;
+	veventq->viommu = viommu;
+	refcount_inc(&viommu->obj.users);
+	veventq->depth = cmd->veventq_depth;
+	list_add_tail(&veventq->node, &viommu->veventqs);
+	veventq->lost_events_header.header.flags =
+		IOMMU_VEVENTQ_FLAG_LOST_EVENTS;
+
+	fdno = iommufd_eventq_init(&veventq->common, "[iommufd-viommu-event]",
+				   ucmd->ictx, &iommufd_veventq_fops);
+	if (fdno < 0) {
+		rc = fdno;
+		goto out_abort;
+	}
+
+	cmd->out_veventq_id = veventq->common.obj.id;
+	cmd->out_veventq_fd = fdno;
+
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+	if (rc)
+		goto out_put_fdno;
+
+	iommufd_object_finalize(ucmd->ictx, &veventq->common.obj);
+	fd_install(fdno, veventq->common.filep);
+	goto out_unlock_veventqs;
+
+out_put_fdno:
+	put_unused_fd(fdno);
+out_abort:
+	iommufd_object_abort_and_destroy(ucmd->ictx, &veventq->common.obj);
+out_unlock_veventqs:
+	up_write(&viommu->veventqs_rwsem);
+	iommufd_put_object(ucmd->ictx, &viommu->obj);
+	return rc;
+}

linux-7.2.3/drivers/iommu/iommufd/hw_pagetable.c

@@ -0,0 +1,563 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES
+ */
+#include <linux/iommu.h>
+#include <uapi/linux/iommufd.h>
+
+#include "../iommu-priv.h"
+#include "iommufd_private.h"
+
+static void __iommufd_hwpt_destroy(struct iommufd_hw_pagetable *hwpt)
+{
+	if (hwpt->domain)
+		iommu_domain_free(hwpt->domain);
+
+	if (hwpt->fault)
+		refcount_dec(&hwpt->fault->common.obj.users);
+}
+
+void iommufd_hwpt_paging_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_hwpt_paging *hwpt_paging =
+		container_of(obj, struct iommufd_hwpt_paging, common.obj);
+
+	if (!list_empty(&hwpt_paging->hwpt_item)) {
+		mutex_lock(&hwpt_paging->ioas->mutex);
+		list_del(&hwpt_paging->hwpt_item);
+		mutex_unlock(&hwpt_paging->ioas->mutex);
+
+		iopt_table_remove_domain(&hwpt_paging->ioas->iopt,
+					 hwpt_paging->common.domain);
+	}
+
+	__iommufd_hwpt_destroy(&hwpt_paging->common);
+	refcount_dec(&hwpt_paging->ioas->obj.users);
+}
+
+void iommufd_hwpt_paging_abort(struct iommufd_object *obj)
+{
+	struct iommufd_hwpt_paging *hwpt_paging =
+		container_of(obj, struct iommufd_hwpt_paging, common.obj);
+
+	/* The ioas->mutex must be held until finalize is called. */
+	lockdep_assert_held(&hwpt_paging->ioas->mutex);
+
+	if (!list_empty(&hwpt_paging->hwpt_item)) {
+		list_del_init(&hwpt_paging->hwpt_item);
+		iopt_table_remove_domain(&hwpt_paging->ioas->iopt,
+					 hwpt_paging->common.domain);
+	}
+	iommufd_hwpt_paging_destroy(obj);
+}
+
+void iommufd_hwpt_nested_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_hwpt_nested *hwpt_nested =
+		container_of(obj, struct iommufd_hwpt_nested, common.obj);
+
+	__iommufd_hwpt_destroy(&hwpt_nested->common);
+	if (hwpt_nested->viommu)
+		refcount_dec(&hwpt_nested->viommu->obj.users);
+	else
+		refcount_dec(&hwpt_nested->parent->common.obj.users);
+}
+
+void iommufd_hwpt_nested_abort(struct iommufd_object *obj)
+{
+	iommufd_hwpt_nested_destroy(obj);
+}
+
+static int
+iommufd_hwpt_paging_enforce_cc(struct iommufd_hwpt_paging *hwpt_paging)
+{
+	struct iommu_domain *paging_domain = hwpt_paging->common.domain;
+
+	if (hwpt_paging->enforce_cache_coherency)
+		return 0;
+
+	if (paging_domain->ops->enforce_cache_coherency)
+		hwpt_paging->enforce_cache_coherency =
+			paging_domain->ops->enforce_cache_coherency(
+				paging_domain);
+	if (!hwpt_paging->enforce_cache_coherency)
+		return -EINVAL;
+	return 0;
+}
+
+/**
+ * iommufd_hwpt_paging_alloc() - Get a PAGING iommu_domain for a device
+ * @ictx: iommufd context
+ * @ioas: IOAS to associate the domain with
+ * @idev: Device to get an iommu_domain for
+ * @pasid: PASID to get an iommu_domain for
+ * @flags: Flags from userspace
+ * @immediate_attach: True if idev should be attached to the hwpt
+ * @user_data: The user provided driver specific data describing the domain to
+ *             create
+ *
+ * Allocate a new iommu_domain and return it as a hw_pagetable. The HWPT
+ * will be linked to the given ioas and upon return the underlying iommu_domain
+ * is fully popoulated.
+ *
+ * The caller must hold the ioas->mutex until after
+ * iommufd_object_abort_and_destroy() or iommufd_object_finalize() is called on
+ * the returned hwpt.
+ */
+struct iommufd_hwpt_paging *
+iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas,
+			  struct iommufd_device *idev, ioasid_t pasid,
+			  u32 flags, bool immediate_attach,
+			  const struct iommu_user_data *user_data)
+{
+	const u32 valid_flags = IOMMU_HWPT_ALLOC_NEST_PARENT |
+				IOMMU_HWPT_ALLOC_DIRTY_TRACKING |
+				IOMMU_HWPT_FAULT_ID_VALID |
+				IOMMU_HWPT_ALLOC_PASID;
+	const struct iommu_ops *ops = dev_iommu_ops(idev->dev);
+	struct iommufd_hwpt_paging *hwpt_paging;
+	struct iommufd_hw_pagetable *hwpt;
+	int rc;
+
+	lockdep_assert_held(&ioas->mutex);
+
+	if ((flags || user_data) && !ops->domain_alloc_paging_flags)
+		return ERR_PTR(-EOPNOTSUPP);
+	if (flags & ~valid_flags)
+		return ERR_PTR(-EOPNOTSUPP);
+	if ((flags & IOMMU_HWPT_ALLOC_DIRTY_TRACKING) &&
+	    !device_iommu_capable(idev->dev, IOMMU_CAP_DIRTY_TRACKING))
+		return ERR_PTR(-EOPNOTSUPP);
+	if ((flags & IOMMU_HWPT_FAULT_ID_VALID) &&
+	    (flags & IOMMU_HWPT_ALLOC_NEST_PARENT))
+		return ERR_PTR(-EOPNOTSUPP);
+
+	hwpt_paging = __iommufd_object_alloc(
+		ictx, hwpt_paging, IOMMUFD_OBJ_HWPT_PAGING, common.obj);
+	if (IS_ERR(hwpt_paging))
+		return ERR_CAST(hwpt_paging);
+	hwpt = &hwpt_paging->common;
+	hwpt->pasid_compat = flags & IOMMU_HWPT_ALLOC_PASID;
+
+	INIT_LIST_HEAD(&hwpt_paging->hwpt_item);
+	/* Pairs with iommufd_hw_pagetable_destroy() */
+	refcount_inc(&ioas->obj.users);
+	hwpt_paging->ioas = ioas;
+	hwpt_paging->nest_parent = flags & IOMMU_HWPT_ALLOC_NEST_PARENT;
+
+	if (ops->domain_alloc_paging_flags) {
+		hwpt->domain = ops->domain_alloc_paging_flags(idev->dev,
+				flags & ~IOMMU_HWPT_FAULT_ID_VALID, user_data);
+		if (IS_ERR(hwpt->domain)) {
+			rc = PTR_ERR(hwpt->domain);
+			hwpt->domain = NULL;
+			goto out_abort;
+		}
+		hwpt->domain->owner = ops;
+	} else {
+		hwpt->domain = iommu_paging_domain_alloc(idev->dev);
+		if (IS_ERR(hwpt->domain)) {
+			rc = PTR_ERR(hwpt->domain);
+			hwpt->domain = NULL;
+			goto out_abort;
+		}
+	}
+	hwpt->domain->iommufd_hwpt = hwpt;
+	hwpt->domain->cookie_type = IOMMU_COOKIE_IOMMUFD;
+
+	/*
+	 * Set the coherency mode before we do iopt_table_add_domain() as some
+	 * iommus have a per-PTE bit that controls it and need to decide before
+	 * doing any maps. It is an iommu driver bug to report
+	 * IOMMU_CAP_ENFORCE_CACHE_COHERENCY but fail enforce_cache_coherency on
+	 * a new domain.
+	 *
+	 * The cache coherency mode must be configured here and unchanged later.
+	 * Note that a HWPT (non-CC) created for a device (non-CC) can be later
+	 * reused by another device (either non-CC or CC). However, A HWPT (CC)
+	 * created for a device (CC) cannot be reused by another device (non-CC)
+	 * but only devices (CC). Instead user space in this case would need to
+	 * allocate a separate HWPT (non-CC).
+	 */
+	if (idev->enforce_cache_coherency) {
+		rc = iommufd_hwpt_paging_enforce_cc(hwpt_paging);
+		if (WARN_ON(rc))
+			goto out_abort;
+	}
+
+	/*
+	 * immediate_attach exists only to accommodate iommu drivers that cannot
+	 * directly allocate a domain. These drivers do not finish creating the
+	 * domain until attach is completed. Thus we must have this call
+	 * sequence. Once those drivers are fixed this should be removed.
+	 */
+	if (immediate_attach) {
+		rc = iommufd_hw_pagetable_attach(hwpt, idev, pasid);
+		if (rc)
+			goto out_abort;
+	}
+
+	rc = iopt_table_add_domain(&ioas->iopt, hwpt->domain);
+	if (rc)
+		goto out_detach;
+	list_add_tail(&hwpt_paging->hwpt_item, &ioas->hwpt_list);
+	return hwpt_paging;
+
+out_detach:
+	if (immediate_attach)
+		iommufd_hw_pagetable_detach(idev, pasid);
+out_abort:
+	iommufd_object_abort_and_destroy(ictx, &hwpt->obj);
+	return ERR_PTR(rc);
+}
+
+/**
+ * iommufd_hwpt_nested_alloc() - Get a NESTED iommu_domain for a device
+ * @ictx: iommufd context
+ * @parent: Parent PAGING-type hwpt to associate the domain with
+ * @idev: Device to get an iommu_domain for
+ * @flags: Flags from userspace
+ * @user_data: user_data pointer. Must be valid
+ *
+ * Allocate a new iommu_domain (must be IOMMU_DOMAIN_NESTED) and return it as
+ * a NESTED hw_pagetable. The given parent PAGING-type hwpt must be capable of
+ * being a parent.
+ */
+static struct iommufd_hwpt_nested *
+iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx,
+			  struct iommufd_hwpt_paging *parent,
+			  struct iommufd_device *idev, u32 flags,
+			  const struct iommu_user_data *user_data)
+{
+	const struct iommu_ops *ops = dev_iommu_ops(idev->dev);
+	struct iommufd_hwpt_nested *hwpt_nested;
+	struct iommufd_hw_pagetable *hwpt;
+	int rc;
+
+	if ((flags & ~(IOMMU_HWPT_FAULT_ID_VALID | IOMMU_HWPT_ALLOC_PASID)) ||
+	    !user_data->len || !ops->domain_alloc_nested)
+		return ERR_PTR(-EOPNOTSUPP);
+	if (parent->auto_domain || !parent->nest_parent ||
+	    parent->common.domain->owner != ops)
+		return ERR_PTR(-EINVAL);
+
+	hwpt_nested = __iommufd_object_alloc(
+		ictx, hwpt_nested, IOMMUFD_OBJ_HWPT_NESTED, common.obj);
+	if (IS_ERR(hwpt_nested))
+		return ERR_CAST(hwpt_nested);
+	hwpt = &hwpt_nested->common;
+	hwpt->pasid_compat = flags & IOMMU_HWPT_ALLOC_PASID;
+
+	refcount_inc(&parent->common.obj.users);
+	hwpt_nested->parent = parent;
+
+	hwpt->domain = ops->domain_alloc_nested(
+		idev->dev, parent->common.domain,
+		flags & ~IOMMU_HWPT_FAULT_ID_VALID, user_data);
+	if (IS_ERR(hwpt->domain)) {
+		rc = PTR_ERR(hwpt->domain);
+		hwpt->domain = NULL;
+		goto out_abort;
+	}
+	hwpt->domain->owner = ops;
+	hwpt->domain->iommufd_hwpt = hwpt;
+	hwpt->domain->cookie_type = IOMMU_COOKIE_IOMMUFD;
+
+	if (WARN_ON_ONCE(hwpt->domain->type != IOMMU_DOMAIN_NESTED)) {
+		rc = -EOPNOTSUPP;
+		goto out_abort;
+	}
+	return hwpt_nested;
+
+out_abort:
+	iommufd_object_abort_and_destroy(ictx, &hwpt->obj);
+	return ERR_PTR(rc);
+}
+
+/**
+ * iommufd_viommu_alloc_hwpt_nested() - Get a hwpt_nested for a vIOMMU
+ * @viommu: vIOMMU ojbect to associate the hwpt_nested/domain with
+ * @flags: Flags from userspace
+ * @user_data: user_data pointer. Must be valid
+ *
+ * Allocate a new IOMMU_DOMAIN_NESTED for a vIOMMU and return it as a NESTED
+ * hw_pagetable.
+ */
+static struct iommufd_hwpt_nested *
+iommufd_viommu_alloc_hwpt_nested(struct iommufd_viommu *viommu, u32 flags,
+				 const struct iommu_user_data *user_data)
+{
+	struct iommufd_hwpt_nested *hwpt_nested;
+	struct iommufd_hw_pagetable *hwpt;
+	int rc;
+
+	if (flags & ~(IOMMU_HWPT_FAULT_ID_VALID | IOMMU_HWPT_ALLOC_PASID))
+		return ERR_PTR(-EOPNOTSUPP);
+	if (!user_data->len)
+		return ERR_PTR(-EOPNOTSUPP);
+	if (!viommu->ops || !viommu->ops->alloc_domain_nested)
+		return ERR_PTR(-EOPNOTSUPP);
+
+	hwpt_nested = __iommufd_object_alloc(
+		viommu->ictx, hwpt_nested, IOMMUFD_OBJ_HWPT_NESTED, common.obj);
+	if (IS_ERR(hwpt_nested))
+		return ERR_CAST(hwpt_nested);
+	hwpt = &hwpt_nested->common;
+	hwpt->pasid_compat = flags & IOMMU_HWPT_ALLOC_PASID;
+
+	hwpt_nested->viommu = viommu;
+	refcount_inc(&viommu->obj.users);
+	hwpt_nested->parent = viommu->hwpt;
+
+	hwpt->domain = viommu->ops->alloc_domain_nested(
+		viommu, flags & ~IOMMU_HWPT_FAULT_ID_VALID, user_data);
+	if (IS_ERR(hwpt->domain)) {
+		rc = PTR_ERR(hwpt->domain);
+		hwpt->domain = NULL;
+		goto out_abort;
+	}
+	hwpt->domain->iommufd_hwpt = hwpt;
+	hwpt->domain->owner = viommu->iommu_dev->ops;
+	hwpt->domain->cookie_type = IOMMU_COOKIE_IOMMUFD;
+
+	if (WARN_ON_ONCE(hwpt->domain->type != IOMMU_DOMAIN_NESTED)) {
+		rc = -EOPNOTSUPP;
+		goto out_abort;
+	}
+	return hwpt_nested;
+
+out_abort:
+	iommufd_object_abort_and_destroy(viommu->ictx, &hwpt->obj);
+	return ERR_PTR(rc);
+}
+
+int iommufd_hwpt_alloc(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_hwpt_alloc *cmd = ucmd->cmd;
+	const struct iommu_user_data user_data = {
+		.type = cmd->data_type,
+		.uptr = u64_to_user_ptr(cmd->data_uptr),
+		.len = cmd->data_len,
+	};
+	struct iommufd_hw_pagetable *hwpt;
+	struct iommufd_ioas *ioas = NULL;
+	struct iommufd_object *pt_obj;
+	struct iommufd_device *idev;
+	int rc;
+
+	if (cmd->__reserved)
+		return -EOPNOTSUPP;
+	if ((cmd->data_type == IOMMU_HWPT_DATA_NONE && cmd->data_len) ||
+	    (cmd->data_type != IOMMU_HWPT_DATA_NONE && !cmd->data_len))
+		return -EINVAL;
+
+	idev = iommufd_get_device(ucmd, cmd->dev_id);
+	if (IS_ERR(idev))
+		return PTR_ERR(idev);
+
+	pt_obj = iommufd_get_object(ucmd->ictx, cmd->pt_id, IOMMUFD_OBJ_ANY);
+	if (IS_ERR(pt_obj)) {
+		rc = -EINVAL;
+		goto out_put_idev;
+	}
+
+	if (pt_obj->type == IOMMUFD_OBJ_IOAS) {
+		struct iommufd_hwpt_paging *hwpt_paging;
+
+		ioas = container_of(pt_obj, struct iommufd_ioas, obj);
+		mutex_lock(&ioas->mutex);
+		hwpt_paging = iommufd_hwpt_paging_alloc(
+			ucmd->ictx, ioas, idev, IOMMU_NO_PASID, cmd->flags,
+			false, user_data.len ? &user_data : NULL);
+		if (IS_ERR(hwpt_paging)) {
+			rc = PTR_ERR(hwpt_paging);
+			goto out_unlock;
+		}
+		hwpt = &hwpt_paging->common;
+	} else if (pt_obj->type == IOMMUFD_OBJ_HWPT_PAGING) {
+		struct iommufd_hwpt_nested *hwpt_nested;
+
+		hwpt_nested = iommufd_hwpt_nested_alloc(
+			ucmd->ictx,
+			container_of(pt_obj, struct iommufd_hwpt_paging,
+				     common.obj),
+			idev, cmd->flags, &user_data);
+		if (IS_ERR(hwpt_nested)) {
+			rc = PTR_ERR(hwpt_nested);
+			goto out_unlock;
+		}
+		hwpt = &hwpt_nested->common;
+	} else if (pt_obj->type == IOMMUFD_OBJ_VIOMMU) {
+		struct iommufd_hwpt_nested *hwpt_nested;
+		struct iommufd_viommu *viommu;
+
+		viommu = container_of(pt_obj, struct iommufd_viommu, obj);
+		if (viommu->iommu_dev != __iommu_get_iommu_dev(idev->dev)) {
+			rc = -EINVAL;
+			goto out_unlock;
+		}
+		hwpt_nested = iommufd_viommu_alloc_hwpt_nested(
+			viommu, cmd->flags, &user_data);
+		if (IS_ERR(hwpt_nested)) {
+			rc = PTR_ERR(hwpt_nested);
+			goto out_unlock;
+		}
+		hwpt = &hwpt_nested->common;
+	} else {
+		rc = -EINVAL;
+		goto out_put_pt;
+	}
+
+	if (cmd->flags & IOMMU_HWPT_FAULT_ID_VALID) {
+		struct iommufd_fault *fault;
+
+		fault = iommufd_get_fault(ucmd, cmd->fault_id);
+		if (IS_ERR(fault)) {
+			rc = PTR_ERR(fault);
+			goto out_hwpt;
+		}
+		hwpt->fault = fault;
+		hwpt->domain->iopf_handler = iommufd_fault_iopf_handler;
+		refcount_inc(&fault->common.obj.users);
+		iommufd_put_object(ucmd->ictx, &fault->common.obj);
+	}
+
+	cmd->out_hwpt_id = hwpt->obj.id;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+	if (rc)
+		goto out_hwpt;
+	iommufd_object_finalize(ucmd->ictx, &hwpt->obj);
+	goto out_unlock;
+
+out_hwpt:
+	iommufd_object_abort_and_destroy(ucmd->ictx, &hwpt->obj);
+out_unlock:
+	if (ioas)
+		mutex_unlock(&ioas->mutex);
+out_put_pt:
+	iommufd_put_object(ucmd->ictx, pt_obj);
+out_put_idev:
+	iommufd_put_object(ucmd->ictx, &idev->obj);
+	return rc;
+}
+
+int iommufd_hwpt_set_dirty_tracking(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_hwpt_set_dirty_tracking *cmd = ucmd->cmd;
+	struct iommufd_hwpt_paging *hwpt_paging;
+	struct iommufd_ioas *ioas;
+	int rc = -EOPNOTSUPP;
+	bool enable;
+
+	if (cmd->flags & ~IOMMU_HWPT_DIRTY_TRACKING_ENABLE)
+		return rc;
+
+	hwpt_paging = iommufd_get_hwpt_paging(ucmd, cmd->hwpt_id);
+	if (IS_ERR(hwpt_paging))
+		return PTR_ERR(hwpt_paging);
+
+	ioas = hwpt_paging->ioas;
+	enable = cmd->flags & IOMMU_HWPT_DIRTY_TRACKING_ENABLE;
+
+	rc = iopt_set_dirty_tracking(&ioas->iopt, hwpt_paging->common.domain,
+				     enable);
+
+	iommufd_put_object(ucmd->ictx, &hwpt_paging->common.obj);
+	return rc;
+}
+
+int iommufd_hwpt_get_dirty_bitmap(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_hwpt_get_dirty_bitmap *cmd = ucmd->cmd;
+	struct iommufd_hwpt_paging *hwpt_paging;
+	struct iommufd_ioas *ioas;
+	int rc = -EOPNOTSUPP;
+
+	if ((cmd->flags & ~(IOMMU_HWPT_GET_DIRTY_BITMAP_NO_CLEAR)) ||
+	    cmd->__reserved)
+		return -EOPNOTSUPP;
+
+	hwpt_paging = iommufd_get_hwpt_paging(ucmd, cmd->hwpt_id);
+	if (IS_ERR(hwpt_paging))
+		return PTR_ERR(hwpt_paging);
+
+	ioas = hwpt_paging->ioas;
+	rc = iopt_read_and_clear_dirty_data(
+		&ioas->iopt, hwpt_paging->common.domain, cmd->flags, cmd);
+
+	iommufd_put_object(ucmd->ictx, &hwpt_paging->common.obj);
+	return rc;
+}
+
+/* An arbitrary entry_num cap, far above any realistic invalidation batch */
+#define IOMMU_HWPT_INVALIDATE_ENTRY_NUM_MAX (1U << 19)
+
+int iommufd_hwpt_invalidate(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_hwpt_invalidate *cmd = ucmd->cmd;
+	struct iommu_user_data_array data_array = {
+		.type = cmd->data_type,
+		.uptr = u64_to_user_ptr(cmd->data_uptr),
+		.entry_len = cmd->entry_len,
+		.entry_num = cmd->entry_num,
+	};
+	struct iommufd_object *pt_obj;
+	u32 done_num = 0;
+	int rc;
+
+	if (cmd->__reserved) {
+		rc = -EOPNOTSUPP;
+		goto out;
+	}
+
+	/*
+	 * Bound entry_num and entry_len so a single call cannot pin the CPU;
+	 * entry_len also caps the copy_struct_from_user() trailing-zero scan.
+	 */
+	if (cmd->entry_num &&
+	    (!cmd->data_uptr || !cmd->entry_len || cmd->entry_len > PAGE_SIZE ||
+	     cmd->entry_num > IOMMU_HWPT_INVALIDATE_ENTRY_NUM_MAX)) {
+		rc = -EINVAL;
+		goto out;
+	}
+
+	pt_obj = iommufd_get_object(ucmd->ictx, cmd->hwpt_id, IOMMUFD_OBJ_ANY);
+	if (IS_ERR(pt_obj)) {
+		rc = PTR_ERR(pt_obj);
+		goto out;
+	}
+	if (pt_obj->type == IOMMUFD_OBJ_HWPT_NESTED) {
+		struct iommufd_hw_pagetable *hwpt =
+			container_of(pt_obj, struct iommufd_hw_pagetable, obj);
+
+		if (!hwpt->domain->ops ||
+		    !hwpt->domain->ops->cache_invalidate_user) {
+			rc = -EOPNOTSUPP;
+			goto out_put_pt;
+		}
+		rc = hwpt->domain->ops->cache_invalidate_user(hwpt->domain,
+							      &data_array);
+	} else if (pt_obj->type == IOMMUFD_OBJ_VIOMMU) {
+		struct iommufd_viommu *viommu =
+			container_of(pt_obj, struct iommufd_viommu, obj);
+
+		if (!viommu->ops || !viommu->ops->cache_invalidate) {
+			rc = -EOPNOTSUPP;
+			goto out_put_pt;
+		}
+		rc = viommu->ops->cache_invalidate(viommu, &data_array);
+	} else {
+		rc = -EINVAL;
+		goto out_put_pt;
+	}
+
+	done_num = data_array.entry_num;
+
+out_put_pt:
+	iommufd_put_object(ucmd->ictx, pt_obj);
+out:
+	cmd->entry_num = done_num;
+	if (iommufd_ucmd_respond(ucmd, sizeof(*cmd)))
+		return -EFAULT;
+	return rc;
+}

linux-7.2.3/drivers/iommu/iommufd/io_pagetable.h

@@ -0,0 +1,312 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES.
+ *
+ */
+#ifndef __IO_PAGETABLE_H
+#define __IO_PAGETABLE_H
+
+#include <linux/dma-buf.h>
+#include <linux/interval_tree.h>
+#include <linux/kref.h>
+#include <linux/mutex.h>
+#include <linux/xarray.h>
+
+#include "iommufd_private.h"
+
+struct iommu_domain;
+
+/*
+ * Each io_pagetable is composed of intervals of areas which cover regions of
+ * the iova that are backed by something. iova not covered by areas is not
+ * populated in the page table. Each area is fully populated with pages.
+ *
+ * iovas are in byte units, but must be iopt->iova_alignment aligned.
+ *
+ * pages can be NULL, this means some other thread is still working on setting
+ * up or tearing down the area. When observed under the write side of the
+ * domain_rwsem a NULL pages must mean the area is still being setup and no
+ * domains are filled.
+ *
+ * storage_domain points at an arbitrary iommu_domain that is holding the PFNs
+ * for this area. It is locked by the pages->mutex. This simplifies the locking
+ * as the pages code can rely on the storage_domain without having to get the
+ * iopt->domains_rwsem.
+ *
+ * The io_pagetable::iova_rwsem protects node
+ * The iopt_pages::mutex protects pages_node
+ * iopt and iommu_prot are immutable
+ * The pages::mutex protects num_accesses
+ */
+struct iopt_area {
+	struct interval_tree_node node;
+	struct interval_tree_node pages_node;
+	struct io_pagetable *iopt;
+	struct iopt_pages *pages;
+	struct iommu_domain *storage_domain;
+	/* How many bytes into the first page the area starts */
+	unsigned int page_offset;
+	/* IOMMU_READ, IOMMU_WRITE, etc */
+	int iommu_prot;
+	bool prevent_access : 1;
+	unsigned int num_accesses;
+	unsigned int num_locks;
+};
+
+struct iopt_allowed {
+	struct interval_tree_node node;
+};
+
+struct iopt_reserved {
+	struct interval_tree_node node;
+	void *owner;
+};
+
+int iopt_area_fill_domains(struct iopt_area *area, struct iopt_pages *pages);
+void iopt_area_unfill_domains(struct iopt_area *area, struct iopt_pages *pages);
+
+int iopt_area_fill_domain(struct iopt_area *area, struct iommu_domain *domain);
+void iopt_area_unfill_domain(struct iopt_area *area, struct iopt_pages *pages,
+			     struct iommu_domain *domain);
+void iopt_area_unmap_domain(struct iopt_area *area,
+			    struct iommu_domain *domain);
+
+int iopt_dmabuf_track_domain(struct iopt_pages *pages, struct iopt_area *area,
+			     struct iommu_domain *domain);
+void iopt_dmabuf_untrack_domain(struct iopt_pages *pages,
+				struct iopt_area *area,
+				struct iommu_domain *domain);
+int iopt_dmabuf_track_all_domains(struct iopt_area *area,
+				  struct iopt_pages *pages);
+void iopt_dmabuf_untrack_all_domains(struct iopt_area *area,
+				     struct iopt_pages *pages);
+
+static inline unsigned long iopt_area_index(struct iopt_area *area)
+{
+	return area->pages_node.start;
+}
+
+static inline unsigned long iopt_area_last_index(struct iopt_area *area)
+{
+	return area->pages_node.last;
+}
+
+static inline unsigned long iopt_area_iova(struct iopt_area *area)
+{
+	return area->node.start;
+}
+
+static inline unsigned long iopt_area_last_iova(struct iopt_area *area)
+{
+	return area->node.last;
+}
+
+static inline size_t iopt_area_length(struct iopt_area *area)
+{
+	return (area->node.last - area->node.start) + 1;
+}
+
+/*
+ * Number of bytes from the start of the iopt_pages that the iova begins.
+ * iopt_area_start_byte() / PAGE_SIZE encodes the starting page index
+ * iopt_area_start_byte() % PAGE_SIZE encodes the offset within that page
+ */
+static inline unsigned long iopt_area_start_byte(struct iopt_area *area,
+						 unsigned long iova)
+{
+	if (IS_ENABLED(CONFIG_IOMMUFD_TEST))
+		WARN_ON(iova < iopt_area_iova(area) ||
+			iova > iopt_area_last_iova(area));
+	return (iova - iopt_area_iova(area)) + area->page_offset +
+	       iopt_area_index(area) * PAGE_SIZE;
+}
+
+static inline unsigned long iopt_area_iova_to_index(struct iopt_area *area,
+						    unsigned long iova)
+{
+	return iopt_area_start_byte(area, iova) / PAGE_SIZE;
+}
+
+#define __make_iopt_iter(name)                                                 \
+	static inline struct iopt_##name *iopt_##name##_iter_first(            \
+		struct io_pagetable *iopt, unsigned long start,                \
+		unsigned long last)                                            \
+	{                                                                      \
+		struct interval_tree_node *node;                               \
+									       \
+		lockdep_assert_held(&iopt->iova_rwsem);                        \
+		node = interval_tree_iter_first(&iopt->name##_itree, start,    \
+						last);                         \
+		if (!node)                                                     \
+			return NULL;                                           \
+		return container_of(node, struct iopt_##name, node);           \
+	}                                                                      \
+	static inline struct iopt_##name *iopt_##name##_iter_next(             \
+		struct iopt_##name *last_node, unsigned long start,            \
+		unsigned long last)                                            \
+	{                                                                      \
+		struct interval_tree_node *node;                               \
+									       \
+		node = interval_tree_iter_next(&last_node->node, start, last); \
+		if (!node)                                                     \
+			return NULL;                                           \
+		return container_of(node, struct iopt_##name, node);           \
+	}
+
+__make_iopt_iter(area)
+__make_iopt_iter(allowed)
+__make_iopt_iter(reserved)
+
+struct iopt_area_contig_iter {
+	unsigned long cur_iova;
+	unsigned long last_iova;
+	struct iopt_area *area;
+};
+struct iopt_area *iopt_area_contig_init(struct iopt_area_contig_iter *iter,
+					struct io_pagetable *iopt,
+					unsigned long iova,
+					unsigned long last_iova);
+struct iopt_area *iopt_area_contig_next(struct iopt_area_contig_iter *iter);
+
+static inline bool iopt_area_contig_done(struct iopt_area_contig_iter *iter)
+{
+	return iter->area && iter->last_iova <= iopt_area_last_iova(iter->area);
+}
+
+/*
+ * Iterate over a contiguous list of areas that span the iova,last_iova range.
+ * The caller must check iopt_area_contig_done() after the loop to see if
+ * contiguous areas existed.
+ */
+#define iopt_for_each_contig_area(iter, area, iopt, iova, last_iova)          \
+	for (area = iopt_area_contig_init(iter, iopt, iova, last_iova); area; \
+	     area = iopt_area_contig_next(iter))
+
+enum {
+	IOPT_PAGES_ACCOUNT_NONE = 0,
+	IOPT_PAGES_ACCOUNT_USER = 1,
+	IOPT_PAGES_ACCOUNT_MM = 2,
+	IOPT_PAGES_ACCOUNT_MODE_NUM = 3,
+};
+
+enum iopt_address_type {
+	IOPT_ADDRESS_USER = 0,
+	IOPT_ADDRESS_FILE,
+	IOPT_ADDRESS_DMABUF,
+};
+
+struct iopt_pages_dmabuf_track {
+	struct iommu_domain *domain;
+	struct iopt_area *area;
+	struct list_head elm;
+};
+
+struct iopt_pages_dmabuf {
+	struct dma_buf_attachment *attach;
+	struct phys_vec phys;
+	/* Always PAGE_SIZE aligned */
+	unsigned long start;
+	struct list_head tracker;
+};
+
+/*
+ * This holds a pinned page list for multiple areas of IO address space. The
+ * pages always originate from a linear chunk of userspace VA. Multiple
+ * io_pagetable's, through their iopt_area's, can share a single iopt_pages
+ * which avoids multi-pinning and double accounting of page consumption.
+ *
+ * indexes in this structure are measured in PAGE_SIZE units, are 0 based from
+ * the start of the uptr and extend to npages. pages are pinned dynamically
+ * according to the intervals in the access_itree and domains_itree, npinned
+ * records the current number of pages pinned.
+ */
+struct iopt_pages {
+	struct kref kref;
+	struct mutex mutex;
+	size_t npages;
+	size_t npinned;
+	size_t last_npinned;
+	struct task_struct *source_task;
+	struct mm_struct *source_mm;
+	struct user_struct *source_user;
+	enum iopt_address_type type;
+	union {
+		void __user *uptr;		/* IOPT_ADDRESS_USER */
+		struct {			/* IOPT_ADDRESS_FILE */
+			struct file *file;
+			unsigned long start;
+		};
+		/* IOPT_ADDRESS_DMABUF */
+		struct iopt_pages_dmabuf dmabuf;
+	};
+	bool writable:1;
+	u8 account_mode;
+
+	struct xarray pinned_pfns;
+	/* Of iopt_pages_access::node */
+	struct rb_root_cached access_itree;
+	/* Of iopt_area::pages_node */
+	struct rb_root_cached domains_itree;
+};
+
+static inline bool iopt_is_dmabuf(struct iopt_pages *pages)
+{
+	if (!IS_ENABLED(CONFIG_DMA_SHARED_BUFFER))
+		return false;
+	return pages->type == IOPT_ADDRESS_DMABUF;
+}
+
+static inline bool iopt_dmabuf_revoked(struct iopt_pages *pages)
+{
+	lockdep_assert_held(&pages->mutex);
+	if (iopt_is_dmabuf(pages))
+		return pages->dmabuf.phys.len == 0;
+	return false;
+}
+
+struct iopt_pages *iopt_alloc_user_pages(void __user *uptr,
+					 unsigned long length, bool writable);
+struct iopt_pages *iopt_alloc_file_pages(struct file *file,
+					 unsigned long start_byte,
+					 unsigned long start,
+					 unsigned long length, bool writable);
+struct iopt_pages *iopt_alloc_dmabuf_pages(struct iommufd_ctx *ictx,
+					   struct dma_buf *dmabuf,
+					   unsigned long start_byte,
+					   unsigned long start,
+					   unsigned long length, bool writable);
+void iopt_release_pages(struct kref *kref);
+static inline void iopt_put_pages(struct iopt_pages *pages)
+{
+	kref_put(&pages->kref, iopt_release_pages);
+}
+
+void iopt_pages_fill_from_xarray(struct iopt_pages *pages, unsigned long start,
+				 unsigned long last, struct page **out_pages);
+int iopt_pages_fill_xarray(struct iopt_pages *pages, unsigned long start,
+			   unsigned long last, struct page **out_pages);
+void iopt_pages_unfill_xarray(struct iopt_pages *pages, unsigned long start,
+			      unsigned long last);
+
+int iopt_area_add_access(struct iopt_area *area, unsigned long start,
+			 unsigned long last, struct page **out_pages,
+			 unsigned int flags, bool lock_area);
+void iopt_area_remove_access(struct iopt_area *area, unsigned long start,
+			     unsigned long last, bool unlock_area);
+int iopt_pages_rw_access(struct iopt_pages *pages, unsigned long start_byte,
+			 void *data, unsigned long length, unsigned int flags);
+
+/*
+ * Each interval represents an active iopt_access_pages(), it acts as an
+ * interval lock that keeps the PFNs pinned and stored in the xarray.
+ */
+struct iopt_pages_access {
+	struct interval_tree_node node;
+	unsigned int users;
+};
+
+struct pfn_reader_user;
+
+int iopt_pages_update_pinned(struct iopt_pages *pages, unsigned long npages,
+			     bool inc, struct pfn_reader_user *user);
+
+#endif

linux-7.2.3/drivers/iommu/iommufd/ioas.c

@@ -0,0 +1,667 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES
+ */
+#include <linux/file.h>
+#include <linux/interval_tree.h>
+#include <linux/iommu.h>
+#include <linux/iommufd.h>
+#include <uapi/linux/iommufd.h>
+
+#include "io_pagetable.h"
+
+void iommufd_ioas_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_ioas *ioas = container_of(obj, struct iommufd_ioas, obj);
+	int rc;
+
+	rc = iopt_unmap_all(&ioas->iopt, NULL);
+	WARN_ON(rc && rc != -ENOENT);
+	iopt_destroy_table(&ioas->iopt);
+	mutex_destroy(&ioas->mutex);
+}
+
+struct iommufd_ioas *iommufd_ioas_alloc(struct iommufd_ctx *ictx)
+{
+	struct iommufd_ioas *ioas;
+
+	ioas = iommufd_object_alloc(ictx, ioas, IOMMUFD_OBJ_IOAS);
+	if (IS_ERR(ioas))
+		return ioas;
+
+	iopt_init_table(&ioas->iopt);
+	INIT_LIST_HEAD(&ioas->hwpt_list);
+	mutex_init(&ioas->mutex);
+	return ioas;
+}
+
+int iommufd_ioas_alloc_ioctl(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_ioas_alloc *cmd = ucmd->cmd;
+	struct iommufd_ioas *ioas;
+	int rc;
+
+	if (cmd->flags)
+		return -EOPNOTSUPP;
+
+	ioas = iommufd_ioas_alloc(ucmd->ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	cmd->out_ioas_id = ioas->obj.id;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+	if (rc)
+		goto out_table;
+
+	down_read(&ucmd->ictx->ioas_creation_lock);
+	iommufd_object_finalize(ucmd->ictx, &ioas->obj);
+	up_read(&ucmd->ictx->ioas_creation_lock);
+	return 0;
+
+out_table:
+	iommufd_object_abort_and_destroy(ucmd->ictx, &ioas->obj);
+	return rc;
+}
+
+int iommufd_ioas_iova_ranges(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_iova_range __user *ranges;
+	struct iommu_ioas_iova_ranges *cmd = ucmd->cmd;
+	struct iommufd_ioas *ioas;
+	struct interval_tree_span_iter span;
+	u32 max_iovas;
+	int rc;
+
+	if (cmd->__reserved)
+		return -EOPNOTSUPP;
+
+	ioas = iommufd_get_ioas(ucmd->ictx, cmd->ioas_id);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	down_read(&ioas->iopt.iova_rwsem);
+	max_iovas = cmd->num_iovas;
+	ranges = u64_to_user_ptr(cmd->allowed_iovas);
+	cmd->num_iovas = 0;
+	cmd->out_iova_alignment = ioas->iopt.iova_alignment;
+	interval_tree_for_each_span(&span, &ioas->iopt.reserved_itree, 0,
+				    ULONG_MAX) {
+		if (!span.is_hole)
+			continue;
+		if (cmd->num_iovas < max_iovas) {
+			struct iommu_iova_range elm = {
+				.start = span.start_hole,
+				.last = span.last_hole,
+			};
+
+			if (copy_to_user(&ranges[cmd->num_iovas], &elm,
+					 sizeof(elm))) {
+				rc = -EFAULT;
+				goto out_put;
+			}
+		}
+		cmd->num_iovas++;
+	}
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+	if (rc)
+		goto out_put;
+	if (cmd->num_iovas > max_iovas)
+		rc = -EMSGSIZE;
+out_put:
+	up_read(&ioas->iopt.iova_rwsem);
+	iommufd_put_object(ucmd->ictx, &ioas->obj);
+	return rc;
+}
+
+static int iommufd_ioas_load_iovas(struct rb_root_cached *itree,
+				   struct iommu_iova_range __user *ranges,
+				   u32 num)
+{
+	u32 i;
+
+	for (i = 0; i != num; i++) {
+		struct iommu_iova_range range;
+		struct iopt_allowed *allowed;
+
+		if (copy_from_user(&range, ranges + i, sizeof(range)))
+			return -EFAULT;
+
+		if (range.start >= range.last)
+			return -EINVAL;
+
+		if (interval_tree_iter_first(itree, range.start, range.last))
+			return -EINVAL;
+
+		allowed = kzalloc_obj(*allowed, GFP_KERNEL_ACCOUNT);
+		if (!allowed)
+			return -ENOMEM;
+		allowed->node.start = range.start;
+		allowed->node.last = range.last;
+
+		interval_tree_insert(&allowed->node, itree);
+	}
+	return 0;
+}
+
+int iommufd_ioas_allow_iovas(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_ioas_allow_iovas *cmd = ucmd->cmd;
+	struct rb_root_cached allowed_iova = RB_ROOT_CACHED;
+	struct interval_tree_node *node;
+	struct iommufd_ioas *ioas;
+	struct io_pagetable *iopt;
+	int rc = 0;
+
+	if (cmd->__reserved)
+		return -EOPNOTSUPP;
+
+	ioas = iommufd_get_ioas(ucmd->ictx, cmd->ioas_id);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+	iopt = &ioas->iopt;
+
+	rc = iommufd_ioas_load_iovas(&allowed_iova,
+				     u64_to_user_ptr(cmd->allowed_iovas),
+				     cmd->num_iovas);
+	if (rc)
+		goto out_free;
+
+	/*
+	 * We want the allowed tree update to be atomic, so we have to keep the
+	 * original nodes around, and keep track of the new nodes as we allocate
+	 * memory for them. The simplest solution is to have a new/old tree and
+	 * then swap new for old. On success we free the old tree, on failure we
+	 * free the new tree.
+	 */
+	rc = iopt_set_allow_iova(iopt, &allowed_iova);
+out_free:
+	while ((node = interval_tree_iter_first(&allowed_iova, 0, ULONG_MAX))) {
+		interval_tree_remove(node, &allowed_iova);
+		kfree(container_of(node, struct iopt_allowed, node));
+	}
+	iommufd_put_object(ucmd->ictx, &ioas->obj);
+	return rc;
+}
+
+static int conv_iommu_prot(u32 map_flags)
+{
+	/*
+	 * We provide no manual cache coherency ioctls to userspace and most
+	 * architectures make the CPU ops for cache flushing privileged.
+	 * Therefore we require the underlying IOMMU to support CPU coherent
+	 * operation. Support for IOMMU_CACHE is enforced by the
+	 * IOMMU_CAP_CACHE_COHERENCY test during bind.
+	 */
+	int iommu_prot = IOMMU_CACHE;
+
+	if (map_flags & IOMMU_IOAS_MAP_WRITEABLE)
+		iommu_prot |= IOMMU_WRITE;
+	if (map_flags & IOMMU_IOAS_MAP_READABLE)
+		iommu_prot |= IOMMU_READ;
+	return iommu_prot;
+}
+
+int iommufd_ioas_map_file(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_ioas_map_file *cmd = ucmd->cmd;
+	unsigned long iova = cmd->iova;
+	struct iommufd_ioas *ioas;
+	unsigned int flags = 0;
+	int rc;
+
+	if (cmd->flags &
+	     ~(IOMMU_IOAS_MAP_FIXED_IOVA | IOMMU_IOAS_MAP_WRITEABLE |
+	       IOMMU_IOAS_MAP_READABLE))
+		return -EOPNOTSUPP;
+
+	if (cmd->iova >= ULONG_MAX || cmd->length >= ULONG_MAX)
+		return -EOVERFLOW;
+
+	if (!(cmd->flags &
+	      (IOMMU_IOAS_MAP_WRITEABLE | IOMMU_IOAS_MAP_READABLE)))
+		return -EINVAL;
+
+	ioas = iommufd_get_ioas(ucmd->ictx, cmd->ioas_id);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	if (!(cmd->flags & IOMMU_IOAS_MAP_FIXED_IOVA))
+		flags = IOPT_ALLOC_IOVA;
+
+	rc = iopt_map_file_pages(ucmd->ictx, &ioas->iopt, &iova, cmd->fd,
+				 cmd->start, cmd->length,
+				 conv_iommu_prot(cmd->flags), flags);
+	if (rc)
+		goto out_put;
+
+	cmd->iova = iova;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+out_put:
+	iommufd_put_object(ucmd->ictx, &ioas->obj);
+	return rc;
+}
+
+int iommufd_ioas_map(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_ioas_map *cmd = ucmd->cmd;
+	unsigned long iova = cmd->iova;
+	struct iommufd_ioas *ioas;
+	unsigned int flags = 0;
+	int rc;
+
+	if ((cmd->flags &
+	     ~(IOMMU_IOAS_MAP_FIXED_IOVA | IOMMU_IOAS_MAP_WRITEABLE |
+	       IOMMU_IOAS_MAP_READABLE)) ||
+	    cmd->__reserved)
+		return -EOPNOTSUPP;
+	if (cmd->iova >= ULONG_MAX || cmd->length >= ULONG_MAX)
+		return -EOVERFLOW;
+
+	if (!(cmd->flags &
+	      (IOMMU_IOAS_MAP_WRITEABLE | IOMMU_IOAS_MAP_READABLE)))
+		return -EINVAL;
+
+	ioas = iommufd_get_ioas(ucmd->ictx, cmd->ioas_id);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	if (!(cmd->flags & IOMMU_IOAS_MAP_FIXED_IOVA))
+		flags = IOPT_ALLOC_IOVA;
+	rc = iopt_map_user_pages(ucmd->ictx, &ioas->iopt, &iova,
+				 u64_to_user_ptr(cmd->user_va), cmd->length,
+				 conv_iommu_prot(cmd->flags), flags);
+	if (rc)
+		goto out_put;
+
+	cmd->iova = iova;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+out_put:
+	iommufd_put_object(ucmd->ictx, &ioas->obj);
+	return rc;
+}
+
+int iommufd_ioas_copy(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_ioas_copy *cmd = ucmd->cmd;
+	struct iommufd_ioas *src_ioas;
+	struct iommufd_ioas *dst_ioas;
+	unsigned int flags = 0;
+	LIST_HEAD(pages_list);
+	unsigned long iova;
+	int rc;
+
+	iommufd_test_syz_conv_iova_id(ucmd, cmd->src_ioas_id, &cmd->src_iova,
+				      &cmd->flags);
+
+	if ((cmd->flags &
+	     ~(IOMMU_IOAS_MAP_FIXED_IOVA | IOMMU_IOAS_MAP_WRITEABLE |
+	       IOMMU_IOAS_MAP_READABLE)))
+		return -EOPNOTSUPP;
+	if (cmd->length >= ULONG_MAX || cmd->src_iova >= ULONG_MAX ||
+	    cmd->dst_iova >= ULONG_MAX)
+		return -EOVERFLOW;
+
+	if (!(cmd->flags &
+	      (IOMMU_IOAS_MAP_WRITEABLE | IOMMU_IOAS_MAP_READABLE)))
+		return -EINVAL;
+
+	src_ioas = iommufd_get_ioas(ucmd->ictx, cmd->src_ioas_id);
+	if (IS_ERR(src_ioas))
+		return PTR_ERR(src_ioas);
+	rc = iopt_get_pages(&src_ioas->iopt, cmd->src_iova, cmd->length,
+			    &pages_list);
+	iommufd_put_object(ucmd->ictx, &src_ioas->obj);
+	if (rc)
+		return rc;
+
+	dst_ioas = iommufd_get_ioas(ucmd->ictx, cmd->dst_ioas_id);
+	if (IS_ERR(dst_ioas)) {
+		rc = PTR_ERR(dst_ioas);
+		goto out_pages;
+	}
+
+	if (!(cmd->flags & IOMMU_IOAS_MAP_FIXED_IOVA))
+		flags = IOPT_ALLOC_IOVA;
+	iova = cmd->dst_iova;
+	rc = iopt_map_pages(&dst_ioas->iopt, &pages_list, cmd->length, &iova,
+			    conv_iommu_prot(cmd->flags), flags);
+	if (rc)
+		goto out_put_dst;
+
+	cmd->dst_iova = iova;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+out_put_dst:
+	iommufd_put_object(ucmd->ictx, &dst_ioas->obj);
+out_pages:
+	iopt_free_pages_list(&pages_list);
+	return rc;
+}
+
+int iommufd_ioas_unmap(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_ioas_unmap *cmd = ucmd->cmd;
+	struct iommufd_ioas *ioas;
+	unsigned long unmapped = 0;
+	int rc;
+
+	ioas = iommufd_get_ioas(ucmd->ictx, cmd->ioas_id);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	if (cmd->iova == 0 && cmd->length == U64_MAX) {
+		rc = iopt_unmap_all(&ioas->iopt, &unmapped);
+		if (rc)
+			goto out_put;
+	} else {
+		if (cmd->iova >= ULONG_MAX || cmd->length >= ULONG_MAX) {
+			rc = -EOVERFLOW;
+			goto out_put;
+		}
+		rc = iopt_unmap_iova(&ioas->iopt, cmd->iova, cmd->length,
+				     &unmapped);
+		if (rc)
+			goto out_put;
+		if (!unmapped) {
+			rc = -ENOENT;
+			goto out_put;
+		}
+	}
+
+	cmd->length = unmapped;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+
+out_put:
+	iommufd_put_object(ucmd->ictx, &ioas->obj);
+	return rc;
+}
+
+static void iommufd_release_all_iova_rwsem(struct iommufd_ctx *ictx,
+					   struct xarray *ioas_list)
+{
+	struct iommufd_ioas *ioas;
+	unsigned long index;
+
+	xa_for_each(ioas_list, index, ioas) {
+		up_write(&ioas->iopt.iova_rwsem);
+		refcount_dec(&ioas->obj.users);
+	}
+	up_write(&ictx->ioas_creation_lock);
+	xa_destroy(ioas_list);
+}
+
+static int iommufd_take_all_iova_rwsem(struct iommufd_ctx *ictx,
+				       struct xarray *ioas_list)
+{
+	struct iommufd_object *obj;
+	unsigned long index;
+	int rc;
+
+	/*
+	 * This is very ugly, it is done instead of adding a lock around
+	 * pages->source_mm, which is a performance path for mdev, we just
+	 * obtain the write side of all the iova_rwsems which also protects the
+	 * pages->source_*. Due to copies we can't know which IOAS could read
+	 * from the pages, so we just lock everything. This is the only place
+	 * locks are nested and they are uniformly taken in ID order.
+	 *
+	 * ioas_creation_lock prevents new IOAS from being installed in the
+	 * xarray while we do this, and also prevents more than one thread from
+	 * holding nested locks.
+	 */
+	down_write(&ictx->ioas_creation_lock);
+	xa_lock(&ictx->objects);
+	xa_for_each(&ictx->objects, index, obj) {
+		struct iommufd_ioas *ioas;
+
+		if (!obj || obj->type != IOMMUFD_OBJ_IOAS)
+			continue;
+
+		if (!refcount_inc_not_zero(&obj->users))
+			continue;
+
+		xa_unlock(&ictx->objects);
+
+		ioas = container_of(obj, struct iommufd_ioas, obj);
+		down_write_nest_lock(&ioas->iopt.iova_rwsem,
+				     &ictx->ioas_creation_lock);
+
+		rc = xa_err(xa_store(ioas_list, index, ioas, GFP_KERNEL));
+		if (rc) {
+			iommufd_release_all_iova_rwsem(ictx, ioas_list);
+			return rc;
+		}
+
+		xa_lock(&ictx->objects);
+	}
+	xa_unlock(&ictx->objects);
+	return 0;
+}
+
+static bool need_charge_update(struct iopt_pages *pages)
+{
+	switch (pages->account_mode) {
+	case IOPT_PAGES_ACCOUNT_NONE:
+		return false;
+	case IOPT_PAGES_ACCOUNT_MM:
+		return pages->source_mm != current->mm;
+	case IOPT_PAGES_ACCOUNT_USER:
+		/*
+		 * Update when mm changes because it also accounts
+		 * in mm->pinned_vm.
+		 */
+		return (pages->source_user != current_user()) ||
+		       (pages->source_mm != current->mm);
+	}
+	return true;
+}
+
+static int charge_current(unsigned long *npinned)
+{
+	struct iopt_pages tmp = {
+		.source_mm = current->mm,
+		.source_task = current->group_leader,
+		.source_user = current_user(),
+	};
+	unsigned int account_mode;
+	int rc;
+
+	for (account_mode = 0; account_mode != IOPT_PAGES_ACCOUNT_MODE_NUM;
+	     account_mode++) {
+		if (!npinned[account_mode])
+			continue;
+
+		tmp.account_mode = account_mode;
+		rc = iopt_pages_update_pinned(&tmp, npinned[account_mode], true,
+					      NULL);
+		if (rc)
+			goto err_undo;
+	}
+	return 0;
+
+err_undo:
+	while (account_mode != 0) {
+		account_mode--;
+		if (!npinned[account_mode])
+			continue;
+		tmp.account_mode = account_mode;
+		iopt_pages_update_pinned(&tmp, npinned[account_mode], false,
+					 NULL);
+	}
+	return rc;
+}
+
+static void change_mm(struct iopt_pages *pages)
+{
+	struct task_struct *old_task = pages->source_task;
+	struct user_struct *old_user = pages->source_user;
+	struct mm_struct *old_mm = pages->source_mm;
+
+	pages->source_mm = current->mm;
+	mmgrab(pages->source_mm);
+	mmdrop(old_mm);
+
+	pages->source_task = current->group_leader;
+	get_task_struct(pages->source_task);
+	put_task_struct(old_task);
+
+	pages->source_user = get_uid(current_user());
+	free_uid(old_user);
+}
+
+#define for_each_ioas_area(_xa, _index, _ioas, _area) \
+	xa_for_each((_xa), (_index), (_ioas)) \
+		for (_area = iopt_area_iter_first(&_ioas->iopt, 0, ULONG_MAX); \
+		     _area; \
+		     _area = iopt_area_iter_next(_area, 0, ULONG_MAX))
+
+int iommufd_ioas_change_process(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_ioas_change_process *cmd = ucmd->cmd;
+	struct iommufd_ctx *ictx = ucmd->ictx;
+	unsigned long all_npinned[IOPT_PAGES_ACCOUNT_MODE_NUM] = {};
+	struct iommufd_ioas *ioas;
+	struct iopt_area *area;
+	struct iopt_pages *pages;
+	struct xarray ioas_list;
+	unsigned long index;
+	int rc;
+
+	if (cmd->__reserved)
+		return -EOPNOTSUPP;
+
+	xa_init(&ioas_list);
+	rc = iommufd_take_all_iova_rwsem(ictx, &ioas_list);
+	if (rc)
+		return rc;
+
+	for_each_ioas_area(&ioas_list, index, ioas, area)  {
+		if (!area->pages) {
+			rc = -EBUSY;
+			goto out;
+		}
+		if (area->pages->type != IOPT_ADDRESS_FILE) {
+			rc = -EINVAL;
+			goto out;
+		}
+	}
+
+	/*
+	 * Count last_pinned pages, then clear it to avoid double counting
+	 * if the same iopt_pages is visited multiple times in this loop.
+	 * Since we are under all the locks, npinned == last_npinned, so we
+	 * can easily restore last_npinned before we return.
+	 */
+	for_each_ioas_area(&ioas_list, index, ioas, area)  {
+		pages = area->pages;
+
+		if (need_charge_update(pages)) {
+			all_npinned[pages->account_mode] += pages->last_npinned;
+			pages->last_npinned = 0;
+		}
+	}
+
+	rc = charge_current(all_npinned);
+
+	if (rc) {
+		/* Charge failed.  Fix last_npinned and bail. */
+		for_each_ioas_area(&ioas_list, index, ioas, area)
+			area->pages->last_npinned = area->pages->npinned;
+		goto out;
+	}
+
+	for_each_ioas_area(&ioas_list, index, ioas, area) {
+		pages = area->pages;
+
+		/* Uncharge the old one (which also restores last_npinned) */
+		if (need_charge_update(pages)) {
+			int r = iopt_pages_update_pinned(pages, pages->npinned,
+							 false, NULL);
+
+			if (WARN_ON(r))
+				rc = r;
+		}
+		change_mm(pages);
+	}
+
+out:
+	iommufd_release_all_iova_rwsem(ictx, &ioas_list);
+	return rc;
+}
+
+int iommufd_option_rlimit_mode(struct iommu_option *cmd,
+			       struct iommufd_ctx *ictx)
+{
+	if (cmd->object_id)
+		return -EOPNOTSUPP;
+
+	if (cmd->op == IOMMU_OPTION_OP_GET) {
+		cmd->val64 = ictx->account_mode == IOPT_PAGES_ACCOUNT_MM;
+		return 0;
+	}
+	if (cmd->op == IOMMU_OPTION_OP_SET) {
+		int rc = 0;
+
+		if (!capable(CAP_SYS_RESOURCE))
+			return -EPERM;
+
+		xa_lock(&ictx->objects);
+		if (!xa_empty(&ictx->objects)) {
+			rc = -EBUSY;
+		} else {
+			if (cmd->val64 == 0)
+				ictx->account_mode = IOPT_PAGES_ACCOUNT_USER;
+			else if (cmd->val64 == 1)
+				ictx->account_mode = IOPT_PAGES_ACCOUNT_MM;
+			else
+				rc = -EINVAL;
+		}
+		xa_unlock(&ictx->objects);
+
+		return rc;
+	}
+	return -EOPNOTSUPP;
+}
+
+static int iommufd_ioas_option_huge_pages(struct iommu_option *cmd,
+					  struct iommufd_ioas *ioas)
+{
+	if (cmd->op == IOMMU_OPTION_OP_GET) {
+		cmd->val64 = !ioas->iopt.disable_large_pages;
+		return 0;
+	}
+	if (cmd->op == IOMMU_OPTION_OP_SET) {
+		if (cmd->val64 == 0)
+			return iopt_disable_large_pages(&ioas->iopt);
+		if (cmd->val64 == 1) {
+			iopt_enable_large_pages(&ioas->iopt);
+			return 0;
+		}
+		return -EINVAL;
+	}
+	return -EOPNOTSUPP;
+}
+
+int iommufd_ioas_option(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_option *cmd = ucmd->cmd;
+	struct iommufd_ioas *ioas;
+	int rc = 0;
+
+	if (cmd->__reserved)
+		return -EOPNOTSUPP;
+
+	ioas = iommufd_get_ioas(ucmd->ictx, cmd->object_id);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	switch (cmd->option_id) {
+	case IOMMU_OPTION_HUGE_PAGES:
+		rc = iommufd_ioas_option_huge_pages(cmd, ioas);
+		break;
+	default:
+		rc = -EOPNOTSUPP;
+	}
+
+	iommufd_put_object(ucmd->ictx, &ioas->obj);
+	return rc;
+}

linux-7.2.3/drivers/iommu/iommufd/iommufd_private.h

@@ -0,0 +1,749 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES
+ */
+#ifndef __IOMMUFD_PRIVATE_H
+#define __IOMMUFD_PRIVATE_H
+
+#include <linux/iommu.h>
+#include <linux/iommufd.h>
+#include <linux/iova_bitmap.h>
+#include <linux/maple_tree.h>
+#include <linux/rwsem.h>
+#include <linux/uaccess.h>
+#include <linux/xarray.h>
+#include <uapi/linux/iommufd.h>
+
+#include "../iommu-priv.h"
+
+struct iommu_domain;
+struct iommu_group;
+struct iommu_option;
+struct iommufd_device;
+struct dma_buf_attachment;
+
+struct iommufd_sw_msi_map {
+	struct list_head sw_msi_item;
+	phys_addr_t sw_msi_start;
+	phys_addr_t msi_addr;
+	unsigned int pgoff;
+	unsigned int id;
+};
+
+/* Bitmap of struct iommufd_sw_msi_map::id */
+struct iommufd_sw_msi_maps {
+	DECLARE_BITMAP(bitmap, 64);
+};
+
+#ifdef CONFIG_IRQ_MSI_IOMMU
+int iommufd_sw_msi_install(struct iommufd_ctx *ictx,
+			   struct iommufd_hwpt_paging *hwpt_paging,
+			   struct iommufd_sw_msi_map *msi_map);
+#endif
+
+struct iommufd_ctx {
+	struct file *file;
+	struct xarray objects;
+	struct xarray groups;
+	wait_queue_head_t destroy_wait;
+	struct rw_semaphore ioas_creation_lock;
+	struct maple_tree mt_mmap;
+
+	struct mutex sw_msi_lock;
+	struct list_head sw_msi_list;
+	unsigned int sw_msi_id;
+
+	u8 account_mode;
+	/* Compatibility with VFIO no iommu */
+	u8 no_iommu_mode;
+	struct iommufd_ioas *vfio_ioas;
+};
+
+/* Entry for iommufd_ctx::mt_mmap */
+struct iommufd_mmap {
+	struct iommufd_object *owner;
+
+	/* Page-shifted start position in mt_mmap to validate vma->vm_pgoff */
+	unsigned long vm_pgoff;
+
+	/* Physical range for io_remap_pfn_range() */
+	phys_addr_t mmio_addr;
+	size_t length;
+};
+
+/*
+ * The IOVA to PFN map. The map automatically copies the PFNs into multiple
+ * domains and permits sharing of PFNs between io_pagetable instances. This
+ * supports both a design where IOAS's are 1:1 with a domain (eg because the
+ * domain is HW customized), or where the IOAS is 1:N with multiple generic
+ * domains.  The io_pagetable holds an interval tree of iopt_areas which point
+ * to shared iopt_pages which hold the pfns mapped to the page table.
+ *
+ * The locking order is domains_rwsem -> iova_rwsem -> pages::mutex
+ */
+struct io_pagetable {
+	struct rw_semaphore domains_rwsem;
+	struct xarray domains;
+	struct xarray access_list;
+	unsigned int next_domain_id;
+
+	struct rw_semaphore iova_rwsem;
+	struct rb_root_cached area_itree;
+	/* IOVA that cannot become reserved, struct iopt_allowed */
+	struct rb_root_cached allowed_itree;
+	/* IOVA that cannot be allocated, struct iopt_reserved */
+	struct rb_root_cached reserved_itree;
+	u8 disable_large_pages;
+	unsigned long iova_alignment;
+};
+
+void iopt_init_table(struct io_pagetable *iopt);
+void iopt_destroy_table(struct io_pagetable *iopt);
+int iopt_get_pages(struct io_pagetable *iopt, unsigned long iova,
+		   unsigned long length, struct list_head *pages_list);
+void iopt_free_pages_list(struct list_head *pages_list);
+enum {
+	IOPT_ALLOC_IOVA = 1 << 0,
+};
+int iopt_map_user_pages(struct iommufd_ctx *ictx, struct io_pagetable *iopt,
+			unsigned long *iova, void __user *uptr,
+			unsigned long length, int iommu_prot,
+			unsigned int flags);
+int iopt_map_file_pages(struct iommufd_ctx *ictx, struct io_pagetable *iopt,
+			unsigned long *iova, int fd,
+			unsigned long start, unsigned long length,
+			int iommu_prot, unsigned int flags);
+int iopt_map_pages(struct io_pagetable *iopt, struct list_head *pages_list,
+		   unsigned long length, unsigned long *dst_iova,
+		   int iommu_prot, unsigned int flags);
+int iopt_unmap_iova(struct io_pagetable *iopt, unsigned long iova,
+		    unsigned long length, unsigned long *unmapped);
+int iopt_unmap_all(struct io_pagetable *iopt, unsigned long *unmapped);
+
+int iopt_read_and_clear_dirty_data(struct io_pagetable *iopt,
+				   struct iommu_domain *domain,
+				   unsigned long flags,
+				   struct iommu_hwpt_get_dirty_bitmap *bitmap);
+int iopt_set_dirty_tracking(struct io_pagetable *iopt,
+			    struct iommu_domain *domain, bool enable);
+
+void iommufd_access_notify_unmap(struct io_pagetable *iopt, unsigned long iova,
+				 unsigned long length);
+int iopt_table_add_domain(struct io_pagetable *iopt,
+			  struct iommu_domain *domain);
+void iopt_table_remove_domain(struct io_pagetable *iopt,
+			      struct iommu_domain *domain);
+int iopt_table_enforce_dev_resv_regions(struct io_pagetable *iopt,
+					struct device *dev,
+					phys_addr_t *sw_msi_start);
+int iopt_set_allow_iova(struct io_pagetable *iopt,
+			struct rb_root_cached *allowed_iova);
+int iopt_reserve_iova(struct io_pagetable *iopt, unsigned long start,
+		      unsigned long last, void *owner);
+void iopt_remove_reserved_iova(struct io_pagetable *iopt, void *owner);
+int iopt_cut_iova(struct io_pagetable *iopt, unsigned long *iovas,
+		  size_t num_iovas);
+void iopt_enable_large_pages(struct io_pagetable *iopt);
+int iopt_disable_large_pages(struct io_pagetable *iopt);
+
+struct iommufd_ucmd {
+	struct iommufd_ctx *ictx;
+	void __user *ubuffer;
+	u32 user_size;
+	void *cmd;
+	struct iommufd_object *new_obj;
+};
+
+int iommufd_vfio_ioctl(struct iommufd_ctx *ictx, unsigned int cmd,
+		       unsigned long arg);
+
+/* Copy the response in ucmd->cmd back to userspace. */
+static inline int iommufd_ucmd_respond(struct iommufd_ucmd *ucmd,
+				       size_t cmd_len)
+{
+	if (copy_to_user(ucmd->ubuffer, ucmd->cmd,
+			 min_t(size_t, ucmd->user_size, cmd_len)))
+		return -EFAULT;
+	return 0;
+}
+
+static inline bool iommufd_lock_obj(struct iommufd_object *obj)
+{
+	if (!refcount_inc_not_zero(&obj->users))
+		return false;
+	if (!refcount_inc_not_zero(&obj->wait_cnt)) {
+		/*
+		 * If the caller doesn't already have a ref on obj this must be
+		 * called under the xa_lock. Otherwise the caller is holding a
+		 * ref on users. Thus it cannot be one before this decrement.
+		 */
+		refcount_dec(&obj->users);
+		return false;
+	}
+	return true;
+}
+
+struct iommufd_object *iommufd_get_object(struct iommufd_ctx *ictx, u32 id,
+					  enum iommufd_object_type type);
+static inline void iommufd_put_object(struct iommufd_ctx *ictx,
+				      struct iommufd_object *obj)
+{
+	/*
+	 * Users first, then wait_cnt so that REMOVE_WAIT never sees a spurious
+	 * !0 users with a 0 wait_cnt.
+	 */
+	refcount_dec(&obj->users);
+	if (refcount_dec_and_test(&obj->wait_cnt))
+		wake_up_interruptible_all(&ictx->destroy_wait);
+}
+
+void iommufd_object_abort(struct iommufd_ctx *ictx, struct iommufd_object *obj);
+void iommufd_object_abort_and_destroy(struct iommufd_ctx *ictx,
+				      struct iommufd_object *obj);
+void iommufd_object_finalize(struct iommufd_ctx *ictx,
+			     struct iommufd_object *obj);
+
+enum {
+	REMOVE_WAIT		= BIT(0),
+	REMOVE_OBJ_TOMBSTONE	= BIT(1),
+};
+int iommufd_object_remove(struct iommufd_ctx *ictx,
+			  struct iommufd_object *to_destroy, u32 id,
+			  unsigned int flags);
+
+/*
+ * The caller holds a users refcount and wants to destroy the object. At this
+ * point the caller has no wait_cnt reference and at least the xarray will be
+ * holding one.
+ */
+static inline void iommufd_object_destroy_user(struct iommufd_ctx *ictx,
+					       struct iommufd_object *obj)
+{
+	int ret;
+
+	ret = iommufd_object_remove(ictx, obj, obj->id, REMOVE_WAIT);
+
+	/*
+	 * If there is a bug and we couldn't destroy the object then we did put
+	 * back the caller's users refcount and will eventually try to free it
+	 * again during close.
+	 */
+	WARN_ON(ret);
+}
+
+/*
+ * Similar to iommufd_object_destroy_user(), except that the object ID is left
+ * reserved/tombstoned.
+ */
+static inline void iommufd_object_tombstone_user(struct iommufd_ctx *ictx,
+						 struct iommufd_object *obj)
+{
+	int ret;
+
+	ret = iommufd_object_remove(ictx, obj, obj->id,
+				    REMOVE_WAIT | REMOVE_OBJ_TOMBSTONE);
+
+	/*
+	 * If there is a bug and we couldn't destroy the object then we did put
+	 * back the caller's users refcount and will eventually try to free it
+	 * again during close.
+	 */
+	WARN_ON(ret);
+}
+
+/*
+ * The HWPT allocated by autodomains is used in possibly many devices and
+ * is automatically destroyed when its refcount reaches zero.
+ *
+ * If userspace uses the HWPT manually, even for a short term, then it will
+ * disrupt this refcounting and the auto-free in the kernel will not work.
+ * Userspace that tries to use the automatically allocated HWPT must be careful
+ * to ensure that it is consistently destroyed, eg by not racing accesses
+ * and by not attaching an automatic HWPT to a device manually.
+ */
+static inline void
+iommufd_object_put_and_try_destroy(struct iommufd_ctx *ictx,
+				   struct iommufd_object *obj)
+{
+	iommufd_object_remove(ictx, obj, obj->id, 0);
+}
+
+/*
+ * Callers of these normal object allocators must call iommufd_object_finalize()
+ * to finalize the object, or call iommufd_object_abort_and_destroy() to revert
+ * the allocation.
+ */
+struct iommufd_object *_iommufd_object_alloc(struct iommufd_ctx *ictx,
+					     size_t size,
+					     enum iommufd_object_type type);
+
+#define __iommufd_object_alloc(ictx, ptr, type, obj)                           \
+	container_of(_iommufd_object_alloc(                                    \
+			     ictx,                                             \
+			     sizeof(*(ptr)) + BUILD_BUG_ON_ZERO(               \
+						      offsetof(typeof(*(ptr)), \
+							       obj) != 0),     \
+			     type),                                            \
+		     typeof(*(ptr)), obj)
+
+#define iommufd_object_alloc(ictx, ptr, type) \
+	__iommufd_object_alloc(ictx, ptr, type, obj)
+
+/*
+ * Callers of these _ucmd allocators should not call iommufd_object_finalize()
+ * or iommufd_object_abort_and_destroy(), as the core automatically does that.
+ */
+struct iommufd_object *
+_iommufd_object_alloc_ucmd(struct iommufd_ucmd *ucmd, size_t size,
+			   enum iommufd_object_type type);
+
+#define __iommufd_object_alloc_ucmd(ucmd, ptr, type, obj)                      \
+	container_of(_iommufd_object_alloc_ucmd(                               \
+			     ucmd,                                             \
+			     sizeof(*(ptr)) + BUILD_BUG_ON_ZERO(               \
+						      offsetof(typeof(*(ptr)), \
+							       obj) != 0),     \
+			     type),                                            \
+		     typeof(*(ptr)), obj)
+
+#define iommufd_object_alloc_ucmd(ucmd, ptr, type) \
+	__iommufd_object_alloc_ucmd(ucmd, ptr, type, obj)
+
+/*
+ * The IO Address Space (IOAS) pagetable is a virtual page table backed by the
+ * io_pagetable object. It is a user controlled mapping of IOVA -> PFNs. The
+ * mapping is copied into all of the associated domains and made available to
+ * in-kernel users.
+ *
+ * Every iommu_domain that is created is wrapped in a iommufd_hw_pagetable
+ * object. When we go to attach a device to an IOAS we need to get an
+ * iommu_domain and wrapping iommufd_hw_pagetable for it.
+ *
+ * An iommu_domain & iommfd_hw_pagetable will be automatically selected
+ * for a device based on the hwpt_list. If no suitable iommu_domain
+ * is found a new iommu_domain will be created.
+ */
+struct iommufd_ioas {
+	struct iommufd_object obj;
+	struct io_pagetable iopt;
+	struct mutex mutex;
+	struct list_head hwpt_list;
+};
+
+static inline struct iommufd_ioas *iommufd_get_ioas(struct iommufd_ctx *ictx,
+						    u32 id)
+{
+	return container_of(iommufd_get_object(ictx, id, IOMMUFD_OBJ_IOAS),
+			    struct iommufd_ioas, obj);
+}
+
+struct iommufd_ioas *iommufd_ioas_alloc(struct iommufd_ctx *ictx);
+int iommufd_ioas_alloc_ioctl(struct iommufd_ucmd *ucmd);
+void iommufd_ioas_destroy(struct iommufd_object *obj);
+int iommufd_ioas_iova_ranges(struct iommufd_ucmd *ucmd);
+int iommufd_ioas_allow_iovas(struct iommufd_ucmd *ucmd);
+int iommufd_ioas_map(struct iommufd_ucmd *ucmd);
+int iommufd_ioas_map_file(struct iommufd_ucmd *ucmd);
+int iommufd_ioas_change_process(struct iommufd_ucmd *ucmd);
+int iommufd_ioas_copy(struct iommufd_ucmd *ucmd);
+int iommufd_ioas_unmap(struct iommufd_ucmd *ucmd);
+int iommufd_ioas_option(struct iommufd_ucmd *ucmd);
+int iommufd_option_rlimit_mode(struct iommu_option *cmd,
+			       struct iommufd_ctx *ictx);
+
+int iommufd_vfio_ioas(struct iommufd_ucmd *ucmd);
+int iommufd_check_iova_range(struct io_pagetable *iopt,
+			     struct iommu_hwpt_get_dirty_bitmap *bitmap);
+
+/*
+ * A HW pagetable is called an iommu_domain inside the kernel. This user object
+ * allows directly creating and inspecting the domains. Domains that have kernel
+ * owned page tables will be associated with an iommufd_ioas that provides the
+ * IOVA to PFN map.
+ */
+struct iommufd_hw_pagetable {
+	struct iommufd_object obj;
+	struct iommu_domain *domain;
+	struct iommufd_fault *fault;
+	bool pasid_compat : 1;
+};
+
+struct iommufd_hwpt_paging {
+	struct iommufd_hw_pagetable common;
+	struct iommufd_ioas *ioas;
+	bool auto_domain : 1;
+	bool enforce_cache_coherency : 1;
+	bool nest_parent : 1;
+	/* Head at iommufd_ioas::hwpt_list */
+	struct list_head hwpt_item;
+	struct iommufd_sw_msi_maps present_sw_msi;
+};
+
+struct iommufd_hwpt_nested {
+	struct iommufd_hw_pagetable common;
+	struct iommufd_hwpt_paging *parent;
+	struct iommufd_viommu *viommu;
+};
+
+static inline bool hwpt_is_paging(struct iommufd_hw_pagetable *hwpt)
+{
+	return hwpt->obj.type == IOMMUFD_OBJ_HWPT_PAGING;
+}
+
+static inline struct iommufd_hwpt_paging *
+to_hwpt_paging(struct iommufd_hw_pagetable *hwpt)
+{
+	return container_of(hwpt, struct iommufd_hwpt_paging, common);
+}
+
+static inline struct iommufd_hwpt_nested *
+to_hwpt_nested(struct iommufd_hw_pagetable *hwpt)
+{
+	return container_of(hwpt, struct iommufd_hwpt_nested, common);
+}
+
+static inline struct iommufd_hwpt_paging *
+find_hwpt_paging(struct iommufd_hw_pagetable *hwpt)
+{
+	switch (hwpt->obj.type) {
+	case IOMMUFD_OBJ_HWPT_PAGING:
+		return to_hwpt_paging(hwpt);
+	case IOMMUFD_OBJ_HWPT_NESTED:
+		return to_hwpt_nested(hwpt)->parent;
+	default:
+		return NULL;
+	}
+}
+
+static inline struct iommufd_hwpt_paging *
+iommufd_get_hwpt_paging(struct iommufd_ucmd *ucmd, u32 id)
+{
+	return container_of(iommufd_get_object(ucmd->ictx, id,
+					       IOMMUFD_OBJ_HWPT_PAGING),
+			    struct iommufd_hwpt_paging, common.obj);
+}
+
+static inline struct iommufd_hw_pagetable *
+iommufd_get_hwpt_nested(struct iommufd_ucmd *ucmd, u32 id)
+{
+	return container_of(iommufd_get_object(ucmd->ictx, id,
+					       IOMMUFD_OBJ_HWPT_NESTED),
+			    struct iommufd_hw_pagetable, obj);
+}
+
+int iommufd_hwpt_set_dirty_tracking(struct iommufd_ucmd *ucmd);
+int iommufd_hwpt_get_dirty_bitmap(struct iommufd_ucmd *ucmd);
+
+struct iommufd_hwpt_paging *
+iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas,
+			  struct iommufd_device *idev, ioasid_t pasid,
+			  u32 flags, bool immediate_attach,
+			  const struct iommu_user_data *user_data);
+int iommufd_hw_pagetable_attach(struct iommufd_hw_pagetable *hwpt,
+				struct iommufd_device *idev, ioasid_t pasid);
+struct iommufd_hw_pagetable *
+iommufd_hw_pagetable_detach(struct iommufd_device *idev, ioasid_t pasid);
+void iommufd_hwpt_paging_destroy(struct iommufd_object *obj);
+void iommufd_hwpt_paging_abort(struct iommufd_object *obj);
+void iommufd_hwpt_nested_destroy(struct iommufd_object *obj);
+void iommufd_hwpt_nested_abort(struct iommufd_object *obj);
+int iommufd_hwpt_alloc(struct iommufd_ucmd *ucmd);
+int iommufd_hwpt_invalidate(struct iommufd_ucmd *ucmd);
+
+static inline void iommufd_hw_pagetable_put(struct iommufd_ctx *ictx,
+					    struct iommufd_hw_pagetable *hwpt)
+{
+	if (hwpt->obj.type == IOMMUFD_OBJ_HWPT_PAGING) {
+		struct iommufd_hwpt_paging *hwpt_paging = to_hwpt_paging(hwpt);
+
+		if (hwpt_paging->auto_domain) {
+			lockdep_assert_not_held(&hwpt_paging->ioas->mutex);
+			iommufd_object_put_and_try_destroy(ictx, &hwpt->obj);
+			return;
+		}
+	}
+	refcount_dec(&hwpt->obj.users);
+}
+
+struct iommufd_attach;
+
+struct iommufd_group {
+	struct kref ref;
+	struct mutex lock;
+	struct iommufd_ctx *ictx;
+	struct iommu_group *group;
+	struct xarray pasid_attach;
+	struct iommufd_sw_msi_maps required_sw_msi;
+	phys_addr_t sw_msi_start;
+};
+
+/*
+ * A iommufd_device object represents the binding relationship between a
+ * consuming driver and the iommufd. These objects are created/destroyed by
+ * external drivers, not by userspace.
+ */
+struct iommufd_device {
+	struct iommufd_object obj;
+	struct iommufd_ctx *ictx;
+	struct iommufd_group *igroup;
+	struct list_head group_item;
+	/* always the physical device */
+	struct device *dev;
+	bool enforce_cache_coherency;
+	struct iommufd_vdevice *vdev;
+	bool destroying;
+};
+
+static inline struct iommufd_device *
+iommufd_get_device(struct iommufd_ucmd *ucmd, u32 id)
+{
+	return container_of(iommufd_get_object(ucmd->ictx, id,
+					       IOMMUFD_OBJ_DEVICE),
+			    struct iommufd_device, obj);
+}
+
+void iommufd_device_pre_destroy(struct iommufd_object *obj);
+void iommufd_device_destroy(struct iommufd_object *obj);
+int iommufd_get_hw_info(struct iommufd_ucmd *ucmd);
+
+struct device *iommufd_global_device(void);
+
+struct iommufd_access {
+	struct iommufd_object obj;
+	struct iommufd_ctx *ictx;
+	struct iommufd_ioas *ioas;
+	struct iommufd_ioas *ioas_unpin;
+	struct mutex ioas_lock;
+	const struct iommufd_access_ops *ops;
+	void *data;
+	unsigned long iova_alignment;
+	u32 iopt_access_list_id;
+};
+
+int iopt_add_access(struct io_pagetable *iopt, struct iommufd_access *access);
+void iopt_remove_access(struct io_pagetable *iopt,
+			struct iommufd_access *access, u32 iopt_access_list_id);
+void iommufd_access_destroy_object(struct iommufd_object *obj);
+
+/* iommufd_access for internal use */
+static inline bool iommufd_access_is_internal(struct iommufd_access *access)
+{
+	return !access->ictx;
+}
+
+struct iommufd_access *iommufd_access_create_internal(struct iommufd_ctx *ictx);
+
+static inline void
+iommufd_access_destroy_internal(struct iommufd_ctx *ictx,
+				struct iommufd_access *access)
+{
+	iommufd_object_destroy_user(ictx, &access->obj);
+}
+
+int iommufd_access_attach_internal(struct iommufd_access *access,
+				   struct iommufd_ioas *ioas);
+
+static inline void iommufd_access_detach_internal(struct iommufd_access *access)
+{
+	iommufd_access_detach(access);
+}
+
+struct iommufd_eventq {
+	struct iommufd_object obj;
+	struct iommufd_ctx *ictx;
+	struct file *filep;
+
+	spinlock_t lock; /* protects the deliver list */
+	struct list_head deliver;
+
+	struct wait_queue_head wait_queue;
+};
+
+struct iommufd_attach_handle {
+	struct iommu_attach_handle handle;
+	struct iommufd_device *idev;
+};
+
+/* Convert an iommu attach handle to iommufd handle. */
+#define to_iommufd_handle(hdl)	container_of(hdl, struct iommufd_attach_handle, handle)
+
+/*
+ * An iommufd_fault object represents an interface to deliver I/O page faults
+ * to the user space. These objects are created/destroyed by the user space and
+ * associated with hardware page table objects during page-table allocation.
+ */
+struct iommufd_fault {
+	struct iommufd_eventq common;
+	struct mutex mutex; /* serializes response flows */
+	struct xarray response;
+};
+
+static inline struct iommufd_fault *
+eventq_to_fault(struct iommufd_eventq *eventq)
+{
+	return container_of(eventq, struct iommufd_fault, common);
+}
+
+static inline struct iommufd_fault *
+iommufd_get_fault(struct iommufd_ucmd *ucmd, u32 id)
+{
+	return container_of(iommufd_get_object(ucmd->ictx, id,
+					       IOMMUFD_OBJ_FAULT),
+			    struct iommufd_fault, common.obj);
+}
+
+int iommufd_fault_alloc(struct iommufd_ucmd *ucmd);
+void iommufd_fault_destroy(struct iommufd_object *obj);
+int iommufd_fault_iopf_handler(struct iopf_group *group);
+void iommufd_auto_response_faults(struct iommufd_hw_pagetable *hwpt,
+				  struct iommufd_attach_handle *handle);
+
+/* An iommufd_vevent represents a vIOMMU event in an iommufd_veventq */
+struct iommufd_vevent {
+	struct iommufd_vevent_header header;
+	struct list_head node; /* for iommufd_eventq::deliver */
+	ssize_t data_len;
+	u8 event_data[] __counted_by(data_len);
+};
+
+#define vevent_for_lost_events_header(vevent) \
+	(vevent->header.flags & IOMMU_VEVENTQ_FLAG_LOST_EVENTS)
+
+/*
+ * An iommufd_veventq object represents an interface to deliver vIOMMU events to
+ * the user space. It is created/destroyed by the user space and associated with
+ * a vIOMMU object during the allocations.
+ */
+struct iommufd_veventq {
+	struct iommufd_eventq common;
+	struct iommufd_viommu *viommu;
+	struct list_head node; /* for iommufd_viommu::veventqs */
+
+	enum iommu_veventq_type type;
+	unsigned int depth;
+
+	/* Use common.lock for protection */
+	u32 num_events;
+	u32 sequence;
+
+	/* Must be last as it ends in a flexible-array member. */
+	struct iommufd_vevent lost_events_header;
+};
+
+static inline struct iommufd_veventq *
+eventq_to_veventq(struct iommufd_eventq *eventq)
+{
+	return container_of(eventq, struct iommufd_veventq, common);
+}
+
+static inline struct iommufd_veventq *
+iommufd_get_veventq(struct iommufd_ucmd *ucmd, u32 id)
+{
+	return container_of(iommufd_get_object(ucmd->ictx, id,
+					       IOMMUFD_OBJ_VEVENTQ),
+			    struct iommufd_veventq, common.obj);
+}
+
+int iommufd_veventq_alloc(struct iommufd_ucmd *ucmd);
+void iommufd_veventq_destroy(struct iommufd_object *obj);
+void iommufd_veventq_abort(struct iommufd_object *obj);
+
+static inline void iommufd_vevent_handler(struct iommufd_veventq *veventq,
+					  struct iommufd_vevent *vevent)
+{
+	struct iommufd_eventq *eventq = &veventq->common;
+
+	lockdep_assert_held(&eventq->lock);
+
+	/*
+	 * Remove the lost_events_header and add the new node at the same time.
+	 * Note the new node can be lost_events_header, for a sequence update.
+	 */
+	if (list_is_last(&veventq->lost_events_header.node, &eventq->deliver))
+		list_del(&veventq->lost_events_header.node);
+	list_add_tail(&vevent->node, &eventq->deliver);
+	vevent->header.sequence = veventq->sequence;
+	veventq->sequence = (veventq->sequence + 1) & INT_MAX;
+
+	wake_up_interruptible(&eventq->wait_queue);
+}
+
+static inline struct iommufd_viommu *
+iommufd_get_viommu(struct iommufd_ucmd *ucmd, u32 id)
+{
+	return container_of(iommufd_get_object(ucmd->ictx, id,
+					       IOMMUFD_OBJ_VIOMMU),
+			    struct iommufd_viommu, obj);
+}
+
+static inline struct iommufd_veventq *
+iommufd_viommu_find_veventq(struct iommufd_viommu *viommu,
+			    enum iommu_veventq_type type)
+{
+	struct iommufd_veventq *veventq, *next;
+
+	lockdep_assert_held(&viommu->veventqs_rwsem);
+
+	list_for_each_entry_safe(veventq, next, &viommu->veventqs, node) {
+		if (veventq->type == type)
+			return veventq;
+	}
+	return NULL;
+}
+
+int iommufd_viommu_alloc_ioctl(struct iommufd_ucmd *ucmd);
+void iommufd_viommu_destroy(struct iommufd_object *obj);
+int iommufd_vdevice_alloc_ioctl(struct iommufd_ucmd *ucmd);
+void iommufd_vdevice_destroy(struct iommufd_object *obj);
+void iommufd_vdevice_abort(struct iommufd_object *obj);
+int iommufd_hw_queue_alloc_ioctl(struct iommufd_ucmd *ucmd);
+void iommufd_hw_queue_destroy(struct iommufd_object *obj);
+
+static inline struct iommufd_vdevice *
+iommufd_get_vdevice(struct iommufd_ctx *ictx, u32 id)
+{
+	return container_of(iommufd_get_object(ictx, id,
+					       IOMMUFD_OBJ_VDEVICE),
+			    struct iommufd_vdevice, obj);
+}
+
+#ifdef CONFIG_IOMMUFD_TEST
+int iommufd_test(struct iommufd_ucmd *ucmd);
+void iommufd_selftest_destroy(struct iommufd_object *obj);
+extern size_t iommufd_test_memory_limit;
+void iommufd_test_syz_conv_iova_id(struct iommufd_ucmd *ucmd,
+				   unsigned int ioas_id, u64 *iova, u32 *flags);
+bool iommufd_should_fail(void);
+int __init iommufd_test_init(void);
+void iommufd_test_exit(void);
+bool iommufd_selftest_is_mock_dev(struct device *dev);
+int iommufd_test_dma_buf_iommufd_map(struct dma_buf_attachment *attachment,
+				     struct phys_vec *phys);
+#else
+static inline void iommufd_test_syz_conv_iova_id(struct iommufd_ucmd *ucmd,
+						 unsigned int ioas_id,
+						 u64 *iova, u32 *flags)
+{
+}
+static inline bool iommufd_should_fail(void)
+{
+	return false;
+}
+static inline int __init iommufd_test_init(void)
+{
+	return 0;
+}
+static inline void iommufd_test_exit(void)
+{
+}
+static inline bool iommufd_selftest_is_mock_dev(struct device *dev)
+{
+	return false;
+}
+static inline int
+iommufd_test_dma_buf_iommufd_map(struct dma_buf_attachment *attachment,
+				 struct phys_vec *phys)
+{
+	return -EOPNOTSUPP;
+}
+#endif
+#endif

linux-7.2.3/drivers/iommu/iommufd/iommufd_test.h

@@ -0,0 +1,293 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES.
+ */
+#ifndef _UAPI_IOMMUFD_TEST_H
+#define _UAPI_IOMMUFD_TEST_H
+
+#include <linux/iommufd.h>
+#include <linux/types.h>
+
+enum {
+	IOMMU_TEST_OP_ADD_RESERVED = 1,
+	IOMMU_TEST_OP_MOCK_DOMAIN,
+	IOMMU_TEST_OP_MD_CHECK_MAP,
+	IOMMU_TEST_OP_MD_CHECK_REFS,
+	IOMMU_TEST_OP_CREATE_ACCESS,
+	IOMMU_TEST_OP_DESTROY_ACCESS_PAGES,
+	IOMMU_TEST_OP_ACCESS_PAGES,
+	IOMMU_TEST_OP_ACCESS_RW,
+	IOMMU_TEST_OP_SET_TEMP_MEMORY_LIMIT,
+	IOMMU_TEST_OP_MOCK_DOMAIN_REPLACE,
+	IOMMU_TEST_OP_ACCESS_REPLACE_IOAS,
+	IOMMU_TEST_OP_MOCK_DOMAIN_FLAGS,
+	IOMMU_TEST_OP_DIRTY,
+	IOMMU_TEST_OP_MD_CHECK_IOTLB,
+	IOMMU_TEST_OP_TRIGGER_IOPF,
+	IOMMU_TEST_OP_DEV_CHECK_CACHE,
+	IOMMU_TEST_OP_TRIGGER_VEVENT,
+	IOMMU_TEST_OP_PASID_ATTACH,
+	IOMMU_TEST_OP_PASID_REPLACE,
+	IOMMU_TEST_OP_PASID_DETACH,
+	IOMMU_TEST_OP_PASID_CHECK_HWPT,
+	IOMMU_TEST_OP_DMABUF_GET,
+	IOMMU_TEST_OP_DMABUF_REVOKE,
+};
+
+enum {
+	MOCK_IOMMUPT_DEFAULT = 0,
+	MOCK_IOMMUPT_HUGE,
+};
+
+/* These values are true for MOCK_IOMMUPT_DEFAULT */
+enum {
+	MOCK_APERTURE_START = 1UL << 24,
+	MOCK_APERTURE_LAST = (1UL << 31) - 1,
+	MOCK_PAGE_SIZE = 2048,
+	MOCK_HUGE_PAGE_SIZE = 512 * MOCK_PAGE_SIZE,
+};
+
+enum {
+	MOCK_FLAGS_ACCESS_WRITE = 1 << 0,
+	MOCK_FLAGS_ACCESS_SYZ = 1 << 16,
+};
+
+enum {
+	MOCK_ACCESS_RW_WRITE = 1 << 0,
+	MOCK_ACCESS_RW_SLOW_PATH = 1 << 2,
+};
+
+enum {
+	MOCK_FLAGS_ACCESS_CREATE_NEEDS_PIN_PAGES = 1 << 0,
+};
+
+enum {
+	MOCK_FLAGS_DEVICE_NO_DIRTY = 1 << 0,
+	MOCK_FLAGS_DEVICE_PASID = 1 << 2,
+};
+
+enum {
+	MOCK_NESTED_DOMAIN_IOTLB_ID_MAX = 3,
+	MOCK_NESTED_DOMAIN_IOTLB_NUM = 4,
+};
+
+enum {
+	MOCK_DEV_CACHE_ID_MAX = 3,
+	MOCK_DEV_CACHE_NUM = 4,
+};
+
+/* Reserved for special pasid replace test */
+#define IOMMU_TEST_PASID_RESERVED 1024
+
+struct iommu_test_cmd {
+	__u32 size;
+	__u32 op;
+	__u32 id;
+	__u32 __reserved;
+	union {
+		struct {
+			__aligned_u64 start;
+			__aligned_u64 length;
+		} add_reserved;
+		struct {
+			__u32 out_stdev_id;
+			__u32 out_hwpt_id;
+			/* out_idev_id is the standard iommufd_bind object */
+			__u32 out_idev_id;
+		} mock_domain;
+		struct {
+			__u32 out_stdev_id;
+			__u32 out_hwpt_id;
+			__u32 out_idev_id;
+			/* Expand mock_domain to set mock device flags */
+			__u32 dev_flags;
+		} mock_domain_flags;
+		struct {
+			__u32 pt_id;
+		} mock_domain_replace;
+		struct {
+			__aligned_u64 iova;
+			__aligned_u64 length;
+			__aligned_u64 uptr;
+		} check_map;
+		struct {
+			__aligned_u64 length;
+			__aligned_u64 uptr;
+			__u32 refs;
+		} check_refs;
+		struct {
+			__u32 out_access_fd;
+			__u32 flags;
+		} create_access;
+		struct {
+			__u32 access_pages_id;
+		} destroy_access_pages;
+		struct {
+			__u32 flags;
+			__u32 out_access_pages_id;
+			__aligned_u64 iova;
+			__aligned_u64 length;
+			__aligned_u64 uptr;
+		} access_pages;
+		struct {
+			__aligned_u64 iova;
+			__aligned_u64 length;
+			__aligned_u64 uptr;
+			__u32 flags;
+		} access_rw;
+		struct {
+			__u32 limit;
+		} memory_limit;
+		struct {
+			__u32 ioas_id;
+		} access_replace_ioas;
+		struct {
+			__u32 flags;
+			__aligned_u64 iova;
+			__aligned_u64 length;
+			__aligned_u64 page_size;
+			__aligned_u64 uptr;
+			__aligned_u64 out_nr_dirty;
+		} dirty;
+		struct {
+			__u32 id;
+			__u32 iotlb;
+		} check_iotlb;
+		struct {
+			__u32 dev_id;
+			__u32 pasid;
+			__u32 grpid;
+			__u32 perm;
+			__u64 addr;
+		} trigger_iopf;
+		struct {
+			__u32 id;
+			__u32 cache;
+		} check_dev_cache;
+		struct {
+			__u32 dev_id;
+		} trigger_vevent;
+		struct {
+			__u32 pasid;
+			__u32 pt_id;
+			/* @id is stdev_id */
+		} pasid_attach;
+		struct {
+			__u32 pasid;
+			__u32 pt_id;
+			/* @id is stdev_id */
+		} pasid_replace;
+		struct {
+			__u32 pasid;
+			/* @id is stdev_id */
+		} pasid_detach;
+		struct {
+			__u32 pasid;
+			__u32 hwpt_id;
+			/* @id is stdev_id */
+		} pasid_check;
+		struct {
+			__u32 length;
+			__u32 open_flags;
+		} dmabuf_get;
+		struct {
+			__s32 dmabuf_fd;
+			__u32 revoked;
+		} dmabuf_revoke;
+	};
+	__u32 last;
+};
+#define IOMMU_TEST_CMD _IO(IOMMUFD_TYPE, IOMMUFD_CMD_BASE + 32)
+
+/* Mock device/iommu PASID width */
+#define MOCK_PASID_WIDTH 20
+
+/* Mock structs for IOMMU_DEVICE_GET_HW_INFO ioctl */
+#define IOMMU_HW_INFO_TYPE_SELFTEST	0xfeedbeef
+#define IOMMU_HW_INFO_SELFTEST_REGVAL	0xdeadbeef
+
+struct iommu_test_hw_info {
+	__u32 flags;
+	__u32 test_reg;
+};
+
+/* Should not be equal to any defined value in enum iommu_hwpt_data_type */
+#define IOMMU_HWPT_DATA_SELFTEST 0xdead
+#define IOMMU_TEST_IOTLB_DEFAULT 0xbadbeef
+#define IOMMU_TEST_DEV_CACHE_DEFAULT 0xbaddad
+
+/**
+ * struct iommu_hwpt_selftest
+ *
+ * @iotlb: default mock iotlb value, IOMMU_TEST_IOTLB_DEFAULT
+ */
+struct iommu_hwpt_selftest {
+	__u32 iotlb;
+	__u32 pagetable_type;
+};
+
+/* Should not be equal to any defined value in enum iommu_hwpt_invalidate_data_type */
+#define IOMMU_HWPT_INVALIDATE_DATA_SELFTEST 0xdeadbeef
+#define IOMMU_HWPT_INVALIDATE_DATA_SELFTEST_INVALID 0xdadbeef
+
+/**
+ * struct iommu_hwpt_invalidate_selftest - Invalidation data for Mock driver
+ *                                         (IOMMU_HWPT_INVALIDATE_DATA_SELFTEST)
+ * @flags: Invalidate flags
+ * @iotlb_id: Invalidate iotlb entry index
+ *
+ * If IOMMU_TEST_INVALIDATE_ALL is set in @flags, @iotlb_id will be ignored
+ */
+struct iommu_hwpt_invalidate_selftest {
+#define IOMMU_TEST_INVALIDATE_FLAG_ALL	(1 << 0)
+	__u32 flags;
+	__u32 iotlb_id;
+};
+
+#define IOMMU_VIOMMU_TYPE_SELFTEST 0xdeadbeef
+
+/**
+ * struct iommu_viommu_selftest - vIOMMU data for Mock driver
+ *                                (IOMMU_VIOMMU_TYPE_SELFTEST)
+ * @in_data: Input random data from user space
+ * @out_data: Output data (matching @in_data) to user space
+ * @out_mmap_offset: The offset argument for mmap syscall
+ * @out_mmap_length: The length argument for mmap syscall
+ *
+ * Simply set @out_data=@in_data for a loopback test
+ */
+struct iommu_viommu_selftest {
+	__u32 in_data;
+	__u32 out_data;
+	__aligned_u64 out_mmap_offset;
+	__aligned_u64 out_mmap_length;
+};
+
+/* Should not be equal to any defined value in enum iommu_viommu_invalidate_data_type */
+#define IOMMU_VIOMMU_INVALIDATE_DATA_SELFTEST 0xdeadbeef
+#define IOMMU_VIOMMU_INVALIDATE_DATA_SELFTEST_INVALID 0xdadbeef
+
+/**
+ * struct iommu_viommu_invalidate_selftest - Invalidation data for Mock VIOMMU
+ *                                        (IOMMU_VIOMMU_INVALIDATE_DATA_SELFTEST)
+ * @flags: Invalidate flags
+ * @cache_id: Invalidate cache entry index
+ *
+ * If IOMMU_TEST_INVALIDATE_ALL is set in @flags, @cache_id will be ignored
+ */
+struct iommu_viommu_invalidate_selftest {
+#define IOMMU_TEST_INVALIDATE_FLAG_ALL (1 << 0)
+	__u32 flags;
+	__u32 vdev_id;
+	__u32 cache_id;
+};
+
+#define IOMMU_VEVENTQ_TYPE_SELFTEST 0xbeefbeef
+
+struct iommu_viommu_event_selftest {
+	__u32 virt_id;
+};
+
+#define IOMMU_HW_QUEUE_TYPE_SELFTEST 0xdeadbeef
+#define IOMMU_TEST_HW_QUEUE_MAX 2
+
+#endif

linux-7.2.3/drivers/iommu/iommufd/iova_bitmap.c

@@ -0,0 +1,446 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2022, Oracle and/or its affiliates.
+ * Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved
+ */
+#include <linux/highmem.h>
+#include <linux/iova_bitmap.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+
+#define BITS_PER_PAGE (PAGE_SIZE * BITS_PER_BYTE)
+
+/*
+ * struct iova_bitmap_map - A bitmap representing an IOVA range
+ *
+ * Main data structure for tracking mapped user pages of bitmap data.
+ *
+ * For example, for something recording dirty IOVAs, it will be provided a
+ * struct iova_bitmap structure, as a general structure for iterating the
+ * total IOVA range. The struct iova_bitmap_map, though, represents the
+ * subset of said IOVA space that is pinned by its parent structure (struct
+ * iova_bitmap).
+ *
+ * The user does not need to exact location of the bits in the bitmap.
+ * From user perspective the only API available is iova_bitmap_set() which
+ * records the IOVA *range* in the bitmap by setting the corresponding
+ * bits.
+ *
+ * The bitmap is an array of u64 whereas each bit represents an IOVA of
+ * range of (1 << pgshift). Thus formula for the bitmap data to be set is:
+ *
+ *   data[(iova / page_size) / 64] & (1ULL << (iova % 64))
+ */
+struct iova_bitmap_map {
+	/* base IOVA representing bit 0 of the first page */
+	unsigned long iova;
+
+	/* mapped length */
+	unsigned long length;
+
+	/* page size order that each bit granules to */
+	unsigned long pgshift;
+
+	/* page offset of the first user page pinned */
+	unsigned long pgoff;
+
+	/* number of pages pinned */
+	unsigned long npages;
+
+	/* pinned pages representing the bitmap data */
+	struct page **pages;
+};
+
+/*
+ * struct iova_bitmap - The IOVA bitmap object
+ *
+ * Main data structure for iterating over the bitmap data.
+ *
+ * Abstracts the pinning work and iterates in IOVA ranges.
+ * It uses a windowing scheme and pins the bitmap in relatively
+ * big ranges e.g.
+ *
+ * The bitmap object uses one base page to store all the pinned pages
+ * pointers related to the bitmap. For sizeof(struct page*) == 8 it stores
+ * 512 struct page pointers which, if the base page size is 4K, it means
+ * 2M of bitmap data is pinned at a time. If the iova_bitmap page size is
+ * also 4K then the range window to iterate is 64G.
+ *
+ * For example iterating on a total IOVA range of 4G..128G, it will walk
+ * through this set of ranges:
+ *
+ *    4G  -  68G-1 (64G)
+ *    68G - 128G-1 (64G)
+ *
+ * An example of the APIs on how to use/iterate over the IOVA bitmap:
+ *
+ *   bitmap = iova_bitmap_alloc(iova, length, page_size, data);
+ *   if (IS_ERR(bitmap))
+ *       return PTR_ERR(bitmap);
+ *
+ *   ret = iova_bitmap_for_each(bitmap, arg, dirty_reporter_fn);
+ *
+ *   iova_bitmap_free(bitmap);
+ *
+ * Each iteration of the @dirty_reporter_fn is called with a unique @iova
+ * and @length argument, indicating the current range available through the
+ * iova_bitmap. The @dirty_reporter_fn uses iova_bitmap_set() to mark dirty
+ * areas (@iova_length) within that provided range, as following:
+ *
+ *   iova_bitmap_set(bitmap, iova, iova_length);
+ *
+ * The internals of the object uses an index @mapped_base_index that indexes
+ * which u64 word of the bitmap is mapped, up to @mapped_total_index.
+ * Those keep being incremented until @mapped_total_index is reached while
+ * mapping up to PAGE_SIZE / sizeof(struct page*) maximum of pages.
+ *
+ * The IOVA bitmap is usually located on what tracks DMA mapped ranges or
+ * some form of IOVA range tracking that co-relates to the user passed
+ * bitmap.
+ */
+struct iova_bitmap {
+	/* IOVA range representing the currently mapped bitmap data */
+	struct iova_bitmap_map mapped;
+
+	/* userspace address of the bitmap */
+	u8 __user *bitmap;
+
+	/* u64 index that @mapped points to */
+	unsigned long mapped_base_index;
+
+	/* how many u64 can we walk in total */
+	unsigned long mapped_total_index;
+
+	/* base IOVA of the whole bitmap */
+	unsigned long iova;
+
+	/* length of the IOVA range for the whole bitmap */
+	size_t length;
+};
+
+/*
+ * Converts a relative IOVA to a bitmap index.
+ * This function provides the index into the u64 array (bitmap::bitmap)
+ * for a given IOVA offset.
+ * Relative IOVA means relative to the bitmap::mapped base IOVA
+ * (stored in mapped::iova). All computations in this file are done using
+ * relative IOVAs and thus avoid an extra subtraction against mapped::iova.
+ * The user API iova_bitmap_set() always uses a regular absolute IOVAs.
+ */
+static unsigned long iova_bitmap_offset_to_index(struct iova_bitmap *bitmap,
+						 unsigned long iova)
+{
+	return (iova >> bitmap->mapped.pgshift) /
+	       BITS_PER_TYPE(*bitmap->bitmap);
+}
+
+/*
+ * Converts a bitmap index to a *relative* IOVA.
+ */
+static unsigned long iova_bitmap_index_to_offset(struct iova_bitmap *bitmap,
+						 unsigned long index)
+{
+	unsigned long pgshift = bitmap->mapped.pgshift;
+
+	return (index * BITS_PER_TYPE(*bitmap->bitmap)) << pgshift;
+}
+
+/*
+ * Returns the base IOVA of the mapped range.
+ */
+static unsigned long iova_bitmap_mapped_iova(struct iova_bitmap *bitmap)
+{
+	unsigned long skip = bitmap->mapped_base_index;
+
+	return bitmap->iova + iova_bitmap_index_to_offset(bitmap, skip);
+}
+
+static unsigned long iova_bitmap_mapped_length(struct iova_bitmap *bitmap);
+
+/*
+ * Pins the bitmap user pages for the current range window.
+ * This is internal to IOVA bitmap and called when advancing the
+ * index (@mapped_base_index) or allocating the bitmap.
+ */
+static int iova_bitmap_get(struct iova_bitmap *bitmap)
+{
+	struct iova_bitmap_map *mapped = &bitmap->mapped;
+	unsigned long npages;
+	u8 __user *addr;
+	long ret;
+
+	/*
+	 * @mapped_base_index is the index of the currently mapped u64 words
+	 * that we have access. Anything before @mapped_base_index is not
+	 * mapped. The range @mapped_base_index .. @mapped_total_index-1 is
+	 * mapped but capped at a maximum number of pages.
+	 */
+	npages = DIV_ROUND_UP((bitmap->mapped_total_index -
+			       bitmap->mapped_base_index) *
+			       sizeof(*bitmap->bitmap), PAGE_SIZE);
+
+	/*
+	 * Bitmap address to be pinned is calculated via pointer arithmetic
+	 * with bitmap u64 word index.
+	 */
+	addr = bitmap->bitmap + bitmap->mapped_base_index;
+
+	/*
+	 * We always cap at max number of 'struct page' a base page can fit.
+	 * This is, for example, on x86 means 2M of bitmap data max.
+	 */
+	npages = min(npages + !!offset_in_page(addr),
+		     PAGE_SIZE / sizeof(struct page *));
+
+	ret = pin_user_pages_fast((unsigned long)addr, npages,
+				  FOLL_WRITE, mapped->pages);
+	if (ret <= 0)
+		return -EFAULT;
+
+	mapped->npages = (unsigned long)ret;
+	/* Base IOVA where @pages point to i.e. bit 0 of the first page */
+	mapped->iova = iova_bitmap_mapped_iova(bitmap);
+
+	/*
+	 * offset of the page where pinned pages bit 0 is located.
+	 * This handles the case where the bitmap is not PAGE_SIZE
+	 * aligned.
+	 */
+	mapped->pgoff = offset_in_page(addr);
+	mapped->length = iova_bitmap_mapped_length(bitmap);
+	return 0;
+}
+
+/*
+ * Unpins the bitmap user pages and clears @npages
+ * (un)pinning is abstracted from API user and it's done when advancing
+ * the index or freeing the bitmap.
+ */
+static void iova_bitmap_put(struct iova_bitmap *bitmap)
+{
+	struct iova_bitmap_map *mapped = &bitmap->mapped;
+
+	if (mapped->npages) {
+		unpin_user_pages(mapped->pages, mapped->npages);
+		mapped->npages = 0;
+	}
+}
+
+/**
+ * iova_bitmap_alloc() - Allocates an IOVA bitmap object
+ * @iova: Start address of the IOVA range
+ * @length: Length of the IOVA range
+ * @page_size: Page size of the IOVA bitmap. It defines what each bit
+ *             granularity represents
+ * @data: Userspace address of the bitmap
+ *
+ * Allocates an IOVA object and initializes all its fields including the
+ * first user pages of @data.
+ *
+ * Return: A pointer to a newly allocated struct iova_bitmap
+ * or ERR_PTR() on error.
+ */
+struct iova_bitmap *iova_bitmap_alloc(unsigned long iova, size_t length,
+				      unsigned long page_size, u64 __user *data)
+{
+	struct iova_bitmap_map *mapped;
+	struct iova_bitmap *bitmap;
+	int rc;
+
+	bitmap = kzalloc_obj(*bitmap);
+	if (!bitmap)
+		return ERR_PTR(-ENOMEM);
+
+	mapped = &bitmap->mapped;
+	mapped->pgshift = __ffs(page_size);
+	bitmap->bitmap = (u8 __user *)data;
+	bitmap->mapped_total_index =
+		iova_bitmap_offset_to_index(bitmap, length - 1) + 1;
+	bitmap->iova = iova;
+	bitmap->length = length;
+	mapped->iova = iova;
+	mapped->pages = (struct page **)__get_free_page(GFP_KERNEL);
+	if (!mapped->pages) {
+		rc = -ENOMEM;
+		goto err;
+	}
+
+	return bitmap;
+
+err:
+	iova_bitmap_free(bitmap);
+	return ERR_PTR(rc);
+}
+EXPORT_SYMBOL_NS_GPL(iova_bitmap_alloc, "IOMMUFD");
+
+/**
+ * iova_bitmap_free() - Frees an IOVA bitmap object
+ * @bitmap: IOVA bitmap to free
+ *
+ * It unpins and releases pages array memory and clears any leftover
+ * state.
+ */
+void iova_bitmap_free(struct iova_bitmap *bitmap)
+{
+	struct iova_bitmap_map *mapped = &bitmap->mapped;
+
+	iova_bitmap_put(bitmap);
+
+	if (mapped->pages) {
+		free_page((unsigned long)mapped->pages);
+		mapped->pages = NULL;
+	}
+
+	kfree(bitmap);
+}
+EXPORT_SYMBOL_NS_GPL(iova_bitmap_free, "IOMMUFD");
+
+/*
+ * Returns the remaining bitmap indexes from mapped_total_index to process for
+ * the currently pinned bitmap pages.
+ */
+static unsigned long iova_bitmap_mapped_remaining(struct iova_bitmap *bitmap)
+{
+	unsigned long remaining, bytes;
+
+	bytes = (bitmap->mapped.npages << PAGE_SHIFT) - bitmap->mapped.pgoff;
+
+	remaining = bitmap->mapped_total_index - bitmap->mapped_base_index;
+	remaining = min_t(unsigned long, remaining,
+			  DIV_ROUND_UP(bytes, sizeof(*bitmap->bitmap)));
+
+	return remaining;
+}
+
+/*
+ * Returns the length of the mapped IOVA range.
+ */
+static unsigned long iova_bitmap_mapped_length(struct iova_bitmap *bitmap)
+{
+	unsigned long max_iova = bitmap->iova + bitmap->length - 1;
+	unsigned long iova = iova_bitmap_mapped_iova(bitmap);
+	unsigned long remaining;
+
+	/*
+	 * iova_bitmap_mapped_remaining() returns a number of indexes which
+	 * when converted to IOVA gives us a max length that the bitmap
+	 * pinned data can cover. Afterwards, that is capped to
+	 * only cover the IOVA range in @bitmap::iova .. @bitmap::length.
+	 */
+	remaining = iova_bitmap_index_to_offset(bitmap,
+			iova_bitmap_mapped_remaining(bitmap));
+
+	if (iova + remaining - 1 > max_iova)
+		remaining -= ((iova + remaining - 1) - max_iova);
+
+	return remaining;
+}
+
+/*
+ * Returns true if [@iova..@iova+@length-1] is part of the mapped IOVA range.
+ */
+static bool iova_bitmap_mapped_range(struct iova_bitmap_map *mapped,
+				     unsigned long iova, size_t length)
+{
+	return mapped->npages &&
+		(iova >= mapped->iova &&
+		 (iova + length - 1) <= (mapped->iova + mapped->length - 1));
+}
+
+/*
+ * Advances to a selected range, releases the current pinned
+ * pages and pins the next set of bitmap pages.
+ * Returns 0 on success or otherwise errno.
+ */
+static int iova_bitmap_advance_to(struct iova_bitmap *bitmap,
+				  unsigned long iova)
+{
+	unsigned long index;
+
+	index = iova_bitmap_offset_to_index(bitmap, iova - bitmap->iova);
+	if (index >= bitmap->mapped_total_index)
+		return -EINVAL;
+	bitmap->mapped_base_index = index;
+
+	iova_bitmap_put(bitmap);
+
+	/* Pin the next set of bitmap pages */
+	return iova_bitmap_get(bitmap);
+}
+
+/**
+ * iova_bitmap_for_each() - Iterates over the bitmap
+ * @bitmap: IOVA bitmap to iterate
+ * @opaque: Additional argument to pass to the callback
+ * @fn: Function that gets called for each IOVA range
+ *
+ * Helper function to iterate over bitmap data representing a portion of IOVA
+ * space. It hides the complexity of iterating bitmaps and translating the
+ * mapped bitmap user pages into IOVA ranges to process.
+ *
+ * Return: 0 on success, and an error on failure either upon
+ * iteration or when the callback returns an error.
+ */
+int iova_bitmap_for_each(struct iova_bitmap *bitmap, void *opaque,
+			 iova_bitmap_fn_t fn)
+{
+	return fn(bitmap, bitmap->iova, bitmap->length, opaque);
+}
+EXPORT_SYMBOL_NS_GPL(iova_bitmap_for_each, "IOMMUFD");
+
+/**
+ * iova_bitmap_set() - Records an IOVA range in bitmap
+ * @bitmap: IOVA bitmap
+ * @iova: IOVA to start
+ * @length: IOVA range length
+ *
+ * Set the bits corresponding to the range [iova .. iova+length-1] in
+ * the user bitmap.
+ *
+ */
+void iova_bitmap_set(struct iova_bitmap *bitmap,
+		     unsigned long iova, size_t length)
+{
+	struct iova_bitmap_map *mapped = &bitmap->mapped;
+	unsigned long cur_bit, last_bit, last_page_idx;
+
+update_indexes:
+	if (unlikely(!iova_bitmap_mapped_range(mapped, iova, length))) {
+		/*
+		 * The attempt to advance the base index to @iova
+		 * may fail if it's out of bounds, or pinning the pages
+		 * returns an error.
+		 */
+		if (iova_bitmap_advance_to(bitmap, iova))
+			return;
+	}
+
+	last_page_idx = mapped->npages - 1;
+	cur_bit = ((iova - mapped->iova) >>
+		mapped->pgshift) + mapped->pgoff * BITS_PER_BYTE;
+	last_bit = (((iova + length - 1) - mapped->iova) >>
+		mapped->pgshift) + mapped->pgoff * BITS_PER_BYTE;
+
+	do {
+		unsigned int page_idx = cur_bit / BITS_PER_PAGE;
+		unsigned int offset = cur_bit % BITS_PER_PAGE;
+		unsigned int nbits = min(BITS_PER_PAGE - offset,
+					 last_bit - cur_bit + 1);
+		void *kaddr;
+
+		if (unlikely(page_idx > last_page_idx)) {
+			unsigned long left =
+				((last_bit - cur_bit + 1) << mapped->pgshift);
+
+			iova += (length - left);
+			length = left;
+			goto update_indexes;
+		}
+
+		kaddr = kmap_local_page(mapped->pages[page_idx]);
+		bitmap_set(kaddr, offset, nbits);
+		kunmap_local(kaddr);
+		cur_bit += nbits;
+	} while (cur_bit <= last_bit);
+}
+EXPORT_SYMBOL_NS_GPL(iova_bitmap_set, "IOMMUFD");

linux-7.2.3/drivers/iommu/iommufd/main.c

@@ -0,0 +1,808 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2021 Intel Corporation
+ * Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES
+ *
+ * iommufd provides control over the IOMMU HW objects created by IOMMU kernel
+ * drivers. IOMMU HW objects revolve around IO page tables that map incoming DMA
+ * addresses (IOVA) to CPU addresses.
+ */
+#define pr_fmt(fmt) "iommufd: " fmt
+
+#include <linux/bug.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/iommufd.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <uapi/linux/iommufd.h>
+
+#include "io_pagetable.h"
+#include "iommufd_private.h"
+#include "iommufd_test.h"
+
+struct iommufd_object_ops {
+	size_t file_offset;
+	void (*pre_destroy)(struct iommufd_object *obj);
+	void (*destroy)(struct iommufd_object *obj);
+	void (*abort)(struct iommufd_object *obj);
+};
+static const struct iommufd_object_ops iommufd_object_ops[];
+static struct miscdevice vfio_misc_dev;
+
+struct iommufd_object *_iommufd_object_alloc(struct iommufd_ctx *ictx,
+					     size_t size,
+					     enum iommufd_object_type type)
+{
+	struct iommufd_object *obj;
+	int rc;
+
+	obj = kzalloc(size, GFP_KERNEL_ACCOUNT);
+	if (!obj)
+		return ERR_PTR(-ENOMEM);
+	obj->type = type;
+	/* Starts out bias'd by 1 until it is removed from the xarray */
+	refcount_set(&obj->wait_cnt, 1);
+	refcount_set(&obj->users, 1);
+
+	/*
+	 * Reserve an ID in the xarray but do not publish the pointer yet since
+	 * the caller hasn't initialized it yet. Once the pointer is published
+	 * in the xarray and visible to other threads we can't reliably destroy
+	 * it anymore, so the caller must complete all errorable operations
+	 * before calling iommufd_object_finalize().
+	 */
+	rc = xa_alloc(&ictx->objects, &obj->id, XA_ZERO_ENTRY, xa_limit_31b,
+		      GFP_KERNEL_ACCOUNT);
+	if (rc)
+		goto out_free;
+	return obj;
+out_free:
+	kfree(obj);
+	return ERR_PTR(rc);
+}
+
+struct iommufd_object *_iommufd_object_alloc_ucmd(struct iommufd_ucmd *ucmd,
+						  size_t size,
+						  enum iommufd_object_type type)
+{
+	struct iommufd_object *new_obj;
+
+	/* Something is coded wrong if this is hit */
+	if (WARN_ON(ucmd->new_obj))
+		return ERR_PTR(-EBUSY);
+
+	/*
+	 * An abort op means that its caller needs to invoke it within a lock in
+	 * the caller. So it doesn't work with _iommufd_object_alloc_ucmd() that
+	 * will invoke the abort op in iommufd_object_abort_and_destroy(), which
+	 * must be outside the caller's lock.
+	 */
+	if (WARN_ON(iommufd_object_ops[type].abort))
+		return ERR_PTR(-EOPNOTSUPP);
+
+	new_obj = _iommufd_object_alloc(ucmd->ictx, size, type);
+	if (IS_ERR(new_obj))
+		return new_obj;
+
+	ucmd->new_obj = new_obj;
+	return new_obj;
+}
+
+/*
+ * Allow concurrent access to the object.
+ *
+ * Once another thread can see the object pointer it can prevent object
+ * destruction. Expect for special kernel-only objects there is no in-kernel way
+ * to reliably destroy a single object. Thus all APIs that are creating objects
+ * must use iommufd_object_abort() to handle their errors and only call
+ * iommufd_object_finalize() once object creation cannot fail.
+ */
+void iommufd_object_finalize(struct iommufd_ctx *ictx,
+			     struct iommufd_object *obj)
+{
+	XA_STATE(xas, &ictx->objects, obj->id);
+	void *old;
+
+	xa_lock(&ictx->objects);
+	old = xas_store(&xas, obj);
+	xa_unlock(&ictx->objects);
+	/* obj->id was returned from xa_alloc() so the xas_store() cannot fail */
+	WARN_ON(old != XA_ZERO_ENTRY);
+}
+
+/* Undo _iommufd_object_alloc() if iommufd_object_finalize() was not called */
+void iommufd_object_abort(struct iommufd_ctx *ictx, struct iommufd_object *obj)
+{
+	XA_STATE(xas, &ictx->objects, obj->id);
+	void *old;
+
+	xa_lock(&ictx->objects);
+	old = xas_store(&xas, NULL);
+	xa_unlock(&ictx->objects);
+	WARN_ON(old != XA_ZERO_ENTRY);
+
+	if (WARN_ON(!refcount_dec_and_test(&obj->users)))
+		return;
+
+	kfree(obj);
+}
+
+/*
+ * Abort an object that has been fully initialized and needs destroy, but has
+ * not been finalized.
+ */
+void iommufd_object_abort_and_destroy(struct iommufd_ctx *ictx,
+				      struct iommufd_object *obj)
+{
+	const struct iommufd_object_ops *ops = &iommufd_object_ops[obj->type];
+
+	if (ops->file_offset) {
+		struct file **filep = ((void *)obj) + ops->file_offset;
+
+		/*
+		 * A file should hold a users refcount while the file is open
+		 * and put it back in its release. The file should hold a
+		 * pointer to obj in their private data. Normal fput() is
+		 * deferred to a workqueue and can get out of order with the
+		 * following kfree(obj). Using the sync version ensures the
+		 * release happens immediately. During abort we require the file
+		 * refcount is one at this point - meaning the object alloc
+		 * function cannot do anything to allow another thread to take a
+		 * refcount prior to a guaranteed success.
+		 */
+		if (*filep)
+			__fput_sync(*filep);
+	}
+
+	if (ops->abort)
+		ops->abort(obj);
+	else
+		ops->destroy(obj);
+	iommufd_object_abort(ictx, obj);
+}
+
+struct iommufd_object *iommufd_get_object(struct iommufd_ctx *ictx, u32 id,
+					  enum iommufd_object_type type)
+{
+	struct iommufd_object *obj;
+
+	if (iommufd_should_fail())
+		return ERR_PTR(-ENOENT);
+
+	xa_lock(&ictx->objects);
+	obj = xa_load(&ictx->objects, id);
+	if (!obj || (type != IOMMUFD_OBJ_ANY && obj->type != type) ||
+	    !iommufd_lock_obj(obj))
+		obj = ERR_PTR(-ENOENT);
+	xa_unlock(&ictx->objects);
+	return obj;
+}
+
+static int iommufd_object_dec_wait(struct iommufd_ctx *ictx,
+				   struct iommufd_object *to_destroy)
+{
+	if (refcount_dec_and_test(&to_destroy->wait_cnt))
+		return 0;
+
+	if (iommufd_object_ops[to_destroy->type].pre_destroy)
+		iommufd_object_ops[to_destroy->type].pre_destroy(to_destroy);
+
+	if (wait_event_timeout(ictx->destroy_wait,
+			       refcount_read(&to_destroy->wait_cnt) == 0,
+			       msecs_to_jiffies(60000)))
+		return 0;
+
+	pr_crit("Time out waiting for iommufd object to become free\n");
+	refcount_inc(&to_destroy->wait_cnt);
+	return -EBUSY;
+}
+
+/*
+ * Remove the given object id from the xarray if the only reference to the
+ * object is held by the xarray.
+ */
+int iommufd_object_remove(struct iommufd_ctx *ictx,
+			  struct iommufd_object *to_destroy, u32 id,
+			  unsigned int flags)
+{
+	struct iommufd_object *obj;
+	XA_STATE(xas, &ictx->objects, id);
+	bool zerod_wait_cnt = false;
+	int ret;
+
+	/*
+	 * The purpose of the wait_cnt is to ensure deterministic destruction
+	 * of objects used by external drivers and destroyed by this function.
+	 * Incrementing this wait_cnt should either be short lived, such as
+	 * during ioctl execution, or be revoked and blocked during
+	 * pre_destroy(), such as vdev holding the idev's refcount.
+	 */
+	if (flags & REMOVE_WAIT) {
+		ret = iommufd_object_dec_wait(ictx, to_destroy);
+		if (ret) {
+			/*
+			 * We have a bug. Put back the callers reference and
+			 * defer cleaning this object until close.
+			 */
+			refcount_dec(&to_destroy->users);
+			return ret;
+		}
+		zerod_wait_cnt = true;
+	}
+
+	xa_lock(&ictx->objects);
+	obj = xas_load(&xas);
+	if (to_destroy) {
+		/*
+		 * If the caller is holding a ref on obj we put it here under
+		 * the spinlock.
+		 */
+		refcount_dec(&obj->users);
+
+		if (WARN_ON(obj != to_destroy)) {
+			ret = -ENOENT;
+			goto err_xa;
+		}
+	} else if (xa_is_zero(obj) || !obj) {
+		ret = -ENOENT;
+		goto err_xa;
+	}
+
+	if (!refcount_dec_if_one(&obj->users)) {
+		ret = -EBUSY;
+		goto err_xa;
+	}
+
+	xas_store(&xas, (flags & REMOVE_OBJ_TOMBSTONE) ? XA_ZERO_ENTRY : NULL);
+	if (ictx->vfio_ioas == container_of(obj, struct iommufd_ioas, obj))
+		ictx->vfio_ioas = NULL;
+	xa_unlock(&ictx->objects);
+
+	/*
+	 * Since users is zero any positive wait_cnt must be racing
+	 * iommufd_put_object(), or we have a bug.
+	 */
+	if (!zerod_wait_cnt) {
+		ret = iommufd_object_dec_wait(ictx, obj);
+		if (WARN_ON(ret))
+			return ret;
+	}
+
+	iommufd_object_ops[obj->type].destroy(obj);
+	kfree(obj);
+	return 0;
+
+err_xa:
+	if (zerod_wait_cnt) {
+		/* Restore the xarray owned reference */
+		refcount_set(&obj->wait_cnt, 1);
+	}
+	xa_unlock(&ictx->objects);
+
+	/* The returned object reference count is zero */
+	return ret;
+}
+
+static int iommufd_destroy(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_destroy *cmd = ucmd->cmd;
+
+	return iommufd_object_remove(ucmd->ictx, NULL, cmd->id, 0);
+}
+
+static int iommufd_fops_open(struct inode *inode, struct file *filp)
+{
+	struct iommufd_ctx *ictx;
+
+	ictx = kzalloc_obj(*ictx, GFP_KERNEL_ACCOUNT);
+	if (!ictx)
+		return -ENOMEM;
+
+	/*
+	 * For compatibility with VFIO when /dev/vfio/vfio is opened we default
+	 * to the same rlimit accounting as vfio uses.
+	 */
+	if (IS_ENABLED(CONFIG_IOMMUFD_VFIO_CONTAINER) &&
+	    filp->private_data == &vfio_misc_dev) {
+		ictx->account_mode = IOPT_PAGES_ACCOUNT_MM;
+		pr_info_once("IOMMUFD is providing /dev/vfio/vfio, not VFIO.\n");
+	}
+
+	init_rwsem(&ictx->ioas_creation_lock);
+	xa_init_flags(&ictx->objects, XA_FLAGS_ALLOC1 | XA_FLAGS_ACCOUNT);
+	xa_init(&ictx->groups);
+	ictx->file = filp;
+	mt_init_flags(&ictx->mt_mmap, MT_FLAGS_ALLOC_RANGE);
+	init_waitqueue_head(&ictx->destroy_wait);
+	mutex_init(&ictx->sw_msi_lock);
+	INIT_LIST_HEAD(&ictx->sw_msi_list);
+	filp->private_data = ictx;
+	return 0;
+}
+
+static int iommufd_fops_release(struct inode *inode, struct file *filp)
+{
+	struct iommufd_ctx *ictx = filp->private_data;
+	struct iommufd_sw_msi_map *next;
+	struct iommufd_sw_msi_map *cur;
+	struct iommufd_object *obj;
+
+	/*
+	 * The objects in the xarray form a graph of "users" counts, and we have
+	 * to destroy them in a depth first manner. Leaf objects will reduce the
+	 * users count of interior objects when they are destroyed.
+	 *
+	 * Repeatedly destroying all the "1 users" leaf objects will progress
+	 * until the entire list is destroyed. If this can't progress then there
+	 * is some bug related to object refcounting.
+	 */
+	while (!xa_empty(&ictx->objects)) {
+		unsigned int destroyed = 0;
+		unsigned long index;
+		bool empty = true;
+
+		/*
+		 * We can't use xa_empty() to end the loop as the tombstones
+		 * are stored as XA_ZERO_ENTRY in the xarray. However
+		 * xa_for_each() automatically converts them to NULL and skips
+		 * them causing xa_empty() to be kept false. Thus once
+		 * xa_for_each() finds no further !NULL entries the loop is
+		 * done.
+		 */
+		xa_for_each(&ictx->objects, index, obj) {
+			empty = false;
+			if (!refcount_dec_if_one(&obj->users))
+				continue;
+
+			destroyed++;
+			xa_erase(&ictx->objects, index);
+			iommufd_object_ops[obj->type].destroy(obj);
+			kfree(obj);
+		}
+
+		if (empty)
+			break;
+
+		/* Bug related to users refcount */
+		if (WARN_ON(!destroyed))
+			break;
+	}
+
+	/*
+	 * There may be some tombstones left over from
+	 * iommufd_object_tombstone_user()
+	 */
+	xa_destroy(&ictx->objects);
+
+	WARN_ON(!xa_empty(&ictx->groups));
+
+	mutex_destroy(&ictx->sw_msi_lock);
+	list_for_each_entry_safe(cur, next, &ictx->sw_msi_list, sw_msi_item)
+		kfree(cur);
+
+	kfree(ictx);
+	return 0;
+}
+
+static int iommufd_option(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_option *cmd = ucmd->cmd;
+	int rc;
+
+	if (cmd->__reserved)
+		return -EOPNOTSUPP;
+
+	switch (cmd->option_id) {
+	case IOMMU_OPTION_RLIMIT_MODE:
+		rc = iommufd_option_rlimit_mode(cmd, ucmd->ictx);
+		break;
+	case IOMMU_OPTION_HUGE_PAGES:
+		rc = iommufd_ioas_option(ucmd);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+	if (rc)
+		return rc;
+	if (copy_to_user(&((struct iommu_option __user *)ucmd->ubuffer)->val64,
+			 &cmd->val64, sizeof(cmd->val64)))
+		return -EFAULT;
+	return 0;
+}
+
+union ucmd_buffer {
+	struct iommu_destroy destroy;
+	struct iommu_fault_alloc fault;
+	struct iommu_hw_info info;
+	struct iommu_hw_queue_alloc hw_queue;
+	struct iommu_hwpt_alloc hwpt;
+	struct iommu_hwpt_get_dirty_bitmap get_dirty_bitmap;
+	struct iommu_hwpt_invalidate cache;
+	struct iommu_hwpt_set_dirty_tracking set_dirty_tracking;
+	struct iommu_ioas_alloc alloc;
+	struct iommu_ioas_allow_iovas allow_iovas;
+	struct iommu_ioas_copy ioas_copy;
+	struct iommu_ioas_iova_ranges iova_ranges;
+	struct iommu_ioas_map map;
+	struct iommu_ioas_unmap unmap;
+	struct iommu_option option;
+	struct iommu_vdevice_alloc vdev;
+	struct iommu_veventq_alloc veventq;
+	struct iommu_vfio_ioas vfio_ioas;
+	struct iommu_viommu_alloc viommu;
+#ifdef CONFIG_IOMMUFD_TEST
+	struct iommu_test_cmd test;
+#endif
+};
+
+struct iommufd_ioctl_op {
+	unsigned int size;
+	unsigned int min_size;
+	unsigned int ioctl_num;
+	int (*execute)(struct iommufd_ucmd *ucmd);
+};
+
+#define IOCTL_OP(_ioctl, _fn, _struct, _last)                                  \
+	[_IOC_NR(_ioctl) - IOMMUFD_CMD_BASE] = {                               \
+		.size = sizeof(_struct) +                                      \
+			BUILD_BUG_ON_ZERO(sizeof(union ucmd_buffer) <          \
+					  sizeof(_struct)),                    \
+		.min_size = offsetofend(_struct, _last),                       \
+		.ioctl_num = _ioctl,                                           \
+		.execute = _fn,                                                \
+	}
+static const struct iommufd_ioctl_op iommufd_ioctl_ops[] = {
+	IOCTL_OP(IOMMU_DESTROY, iommufd_destroy, struct iommu_destroy, id),
+	IOCTL_OP(IOMMU_FAULT_QUEUE_ALLOC, iommufd_fault_alloc,
+		 struct iommu_fault_alloc, out_fault_fd),
+	IOCTL_OP(IOMMU_GET_HW_INFO, iommufd_get_hw_info, struct iommu_hw_info,
+		 __reserved),
+	IOCTL_OP(IOMMU_HW_QUEUE_ALLOC, iommufd_hw_queue_alloc_ioctl,
+		 struct iommu_hw_queue_alloc, length),
+	IOCTL_OP(IOMMU_HWPT_ALLOC, iommufd_hwpt_alloc, struct iommu_hwpt_alloc,
+		 __reserved),
+	IOCTL_OP(IOMMU_HWPT_GET_DIRTY_BITMAP, iommufd_hwpt_get_dirty_bitmap,
+		 struct iommu_hwpt_get_dirty_bitmap, data),
+	IOCTL_OP(IOMMU_HWPT_INVALIDATE, iommufd_hwpt_invalidate,
+		 struct iommu_hwpt_invalidate, __reserved),
+	IOCTL_OP(IOMMU_HWPT_SET_DIRTY_TRACKING, iommufd_hwpt_set_dirty_tracking,
+		 struct iommu_hwpt_set_dirty_tracking, __reserved),
+	IOCTL_OP(IOMMU_IOAS_ALLOC, iommufd_ioas_alloc_ioctl,
+		 struct iommu_ioas_alloc, out_ioas_id),
+	IOCTL_OP(IOMMU_IOAS_ALLOW_IOVAS, iommufd_ioas_allow_iovas,
+		 struct iommu_ioas_allow_iovas, allowed_iovas),
+	IOCTL_OP(IOMMU_IOAS_CHANGE_PROCESS, iommufd_ioas_change_process,
+		 struct iommu_ioas_change_process, __reserved),
+	IOCTL_OP(IOMMU_IOAS_COPY, iommufd_ioas_copy, struct iommu_ioas_copy,
+		 src_iova),
+	IOCTL_OP(IOMMU_IOAS_IOVA_RANGES, iommufd_ioas_iova_ranges,
+		 struct iommu_ioas_iova_ranges, out_iova_alignment),
+	IOCTL_OP(IOMMU_IOAS_MAP, iommufd_ioas_map, struct iommu_ioas_map, iova),
+	IOCTL_OP(IOMMU_IOAS_MAP_FILE, iommufd_ioas_map_file,
+		 struct iommu_ioas_map_file, iova),
+	IOCTL_OP(IOMMU_IOAS_UNMAP, iommufd_ioas_unmap, struct iommu_ioas_unmap,
+		 length),
+	IOCTL_OP(IOMMU_OPTION, iommufd_option, struct iommu_option, val64),
+	IOCTL_OP(IOMMU_VDEVICE_ALLOC, iommufd_vdevice_alloc_ioctl,
+		 struct iommu_vdevice_alloc, virt_id),
+	IOCTL_OP(IOMMU_VEVENTQ_ALLOC, iommufd_veventq_alloc,
+		 struct iommu_veventq_alloc, out_veventq_fd),
+	IOCTL_OP(IOMMU_VFIO_IOAS, iommufd_vfio_ioas, struct iommu_vfio_ioas,
+		 __reserved),
+	IOCTL_OP(IOMMU_VIOMMU_ALLOC, iommufd_viommu_alloc_ioctl,
+		 struct iommu_viommu_alloc, out_viommu_id),
+#ifdef CONFIG_IOMMUFD_TEST
+	IOCTL_OP(IOMMU_TEST_CMD, iommufd_test, struct iommu_test_cmd, last),
+#endif
+};
+
+static long iommufd_fops_ioctl(struct file *filp, unsigned int cmd,
+			       unsigned long arg)
+{
+	struct iommufd_ctx *ictx = filp->private_data;
+	const struct iommufd_ioctl_op *op;
+	struct iommufd_ucmd ucmd = {};
+	union ucmd_buffer buf;
+	unsigned int nr;
+	int ret;
+
+	nr = _IOC_NR(cmd);
+	if (nr < IOMMUFD_CMD_BASE ||
+	    (nr - IOMMUFD_CMD_BASE) >= ARRAY_SIZE(iommufd_ioctl_ops))
+		return iommufd_vfio_ioctl(ictx, cmd, arg);
+
+	ucmd.ictx = ictx;
+	ucmd.ubuffer = (void __user *)arg;
+	ret = get_user(ucmd.user_size, (u32 __user *)ucmd.ubuffer);
+	if (ret)
+		return ret;
+
+	op = &iommufd_ioctl_ops[nr - IOMMUFD_CMD_BASE];
+	if (op->ioctl_num != cmd)
+		return -ENOIOCTLCMD;
+	if (ucmd.user_size < op->min_size)
+		return -EINVAL;
+
+	ucmd.cmd = &buf;
+	ret = copy_struct_from_user(ucmd.cmd, op->size, ucmd.ubuffer,
+				    ucmd.user_size);
+	if (ret)
+		return ret;
+	ret = op->execute(&ucmd);
+
+	if (ucmd.new_obj) {
+		if (ret)
+			iommufd_object_abort_and_destroy(ictx, ucmd.new_obj);
+		else
+			iommufd_object_finalize(ictx, ucmd.new_obj);
+	}
+	return ret;
+}
+
+static void iommufd_fops_vma_open(struct vm_area_struct *vma)
+{
+	struct iommufd_mmap *immap = vma->vm_private_data;
+
+	refcount_inc(&immap->owner->users);
+}
+
+static void iommufd_fops_vma_close(struct vm_area_struct *vma)
+{
+	struct iommufd_mmap *immap = vma->vm_private_data;
+
+	refcount_dec(&immap->owner->users);
+}
+
+static const struct vm_operations_struct iommufd_vma_ops = {
+	.open = iommufd_fops_vma_open,
+	.close = iommufd_fops_vma_close,
+};
+
+/* The vm_pgoff must be pre-allocated from mt_mmap, and given to user space */
+static int iommufd_fops_mmap(struct file *filp, struct vm_area_struct *vma)
+{
+	struct iommufd_ctx *ictx = filp->private_data;
+	size_t length = vma->vm_end - vma->vm_start;
+	struct iommufd_mmap *immap;
+	int rc;
+
+	if (!PAGE_ALIGNED(length))
+		return -EINVAL;
+	if (!(vma->vm_flags & VM_SHARED))
+		return -EINVAL;
+	if (vma->vm_flags & VM_EXEC)
+		return -EPERM;
+
+	mtree_lock(&ictx->mt_mmap);
+	/* vma->vm_pgoff carries a page-shifted start position to an immap */
+	immap = mtree_load(&ictx->mt_mmap, vma->vm_pgoff << PAGE_SHIFT);
+	if (!immap || !refcount_inc_not_zero(&immap->owner->users)) {
+		mtree_unlock(&ictx->mt_mmap);
+		return -ENXIO;
+	}
+	mtree_unlock(&ictx->mt_mmap);
+
+	/*
+	 * mtree_load() returns the immap for any contained mmio_addr, so only
+	 * allow the exact immap thing to be mapped
+	 */
+	if (vma->vm_pgoff != immap->vm_pgoff || length != immap->length) {
+		rc = -ENXIO;
+		goto err_refcount;
+	}
+
+	vma->vm_pgoff = 0;
+	vma->vm_private_data = immap;
+	vma->vm_ops = &iommufd_vma_ops;
+	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+
+	rc = io_remap_pfn_range(vma, vma->vm_start,
+				immap->mmio_addr >> PAGE_SHIFT, length,
+				vma->vm_page_prot);
+	if (rc)
+		goto err_refcount;
+	return 0;
+
+err_refcount:
+	refcount_dec(&immap->owner->users);
+	return rc;
+}
+
+static const struct file_operations iommufd_fops = {
+	.owner = THIS_MODULE,
+	.open = iommufd_fops_open,
+	.release = iommufd_fops_release,
+	.unlocked_ioctl = iommufd_fops_ioctl,
+	.mmap = iommufd_fops_mmap,
+};
+
+/**
+ * iommufd_ctx_get - Get a context reference
+ * @ictx: Context to get
+ *
+ * The caller must already hold a valid reference to ictx.
+ */
+void iommufd_ctx_get(struct iommufd_ctx *ictx)
+{
+	get_file(ictx->file);
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_ctx_get, "IOMMUFD");
+
+/**
+ * iommufd_ctx_from_file - Acquires a reference to the iommufd context
+ * @file: File to obtain the reference from
+ *
+ * Returns a pointer to the iommufd_ctx, otherwise ERR_PTR. The struct file
+ * remains owned by the caller and the caller must still do fput. On success
+ * the caller is responsible to call iommufd_ctx_put().
+ */
+struct iommufd_ctx *iommufd_ctx_from_file(struct file *file)
+{
+	struct iommufd_ctx *ictx;
+
+	if (file->f_op != &iommufd_fops)
+		return ERR_PTR(-EBADFD);
+	ictx = file->private_data;
+	iommufd_ctx_get(ictx);
+	return ictx;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_ctx_from_file, "IOMMUFD");
+
+/**
+ * iommufd_ctx_from_fd - Acquires a reference to the iommufd context
+ * @fd: File descriptor to obtain the reference from
+ *
+ * Returns a pointer to the iommufd_ctx, otherwise ERR_PTR. On success
+ * the caller is responsible to call iommufd_ctx_put().
+ */
+struct iommufd_ctx *iommufd_ctx_from_fd(int fd)
+{
+	struct file *file;
+
+	file = fget(fd);
+	if (!file)
+		return ERR_PTR(-EBADF);
+
+	if (file->f_op != &iommufd_fops) {
+		fput(file);
+		return ERR_PTR(-EBADFD);
+	}
+	/* fget is the same as iommufd_ctx_get() */
+	return file->private_data;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_ctx_from_fd, "IOMMUFD");
+
+/**
+ * iommufd_ctx_put - Put back a reference
+ * @ictx: Context to put back
+ */
+void iommufd_ctx_put(struct iommufd_ctx *ictx)
+{
+	fput(ictx->file);
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_ctx_put, "IOMMUFD");
+
+#define IOMMUFD_FILE_OFFSET(_struct, _filep, _obj)                           \
+	.file_offset = (offsetof(_struct, _filep) +                          \
+			BUILD_BUG_ON_ZERO(!__same_type(                      \
+				struct file *, ((_struct *)NULL)->_filep)) + \
+			BUILD_BUG_ON_ZERO(offsetof(_struct, _obj)))
+
+static const struct iommufd_object_ops iommufd_object_ops[] = {
+	[IOMMUFD_OBJ_ACCESS] = {
+		.destroy = iommufd_access_destroy_object,
+	},
+	[IOMMUFD_OBJ_DEVICE] = {
+		.pre_destroy = iommufd_device_pre_destroy,
+		.destroy = iommufd_device_destroy,
+	},
+	[IOMMUFD_OBJ_FAULT] = {
+		.destroy = iommufd_fault_destroy,
+		IOMMUFD_FILE_OFFSET(struct iommufd_fault, common.filep, common.obj),
+	},
+	[IOMMUFD_OBJ_HW_QUEUE] = {
+		.destroy = iommufd_hw_queue_destroy,
+	},
+	[IOMMUFD_OBJ_HWPT_PAGING] = {
+		.destroy = iommufd_hwpt_paging_destroy,
+		.abort = iommufd_hwpt_paging_abort,
+	},
+	[IOMMUFD_OBJ_HWPT_NESTED] = {
+		.destroy = iommufd_hwpt_nested_destroy,
+		.abort = iommufd_hwpt_nested_abort,
+	},
+	[IOMMUFD_OBJ_IOAS] = {
+		.destroy = iommufd_ioas_destroy,
+	},
+	[IOMMUFD_OBJ_VDEVICE] = {
+		.destroy = iommufd_vdevice_destroy,
+		.abort = iommufd_vdevice_abort,
+	},
+	[IOMMUFD_OBJ_VEVENTQ] = {
+		.destroy = iommufd_veventq_destroy,
+		.abort = iommufd_veventq_abort,
+		IOMMUFD_FILE_OFFSET(struct iommufd_veventq, common.filep, common.obj),
+	},
+	[IOMMUFD_OBJ_VIOMMU] = {
+		.destroy = iommufd_viommu_destroy,
+	},
+#ifdef CONFIG_IOMMUFD_TEST
+	[IOMMUFD_OBJ_SELFTEST] = {
+		.destroy = iommufd_selftest_destroy,
+	},
+#endif
+};
+
+static struct miscdevice iommu_misc_dev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "iommu",
+	.fops = &iommufd_fops,
+	.nodename = "iommu",
+	.mode = 0660,
+};
+
+static struct miscdevice vfio_misc_dev = {
+	.minor = VFIO_MINOR,
+	.name = "vfio",
+	.fops = &iommufd_fops,
+	.nodename = "vfio/vfio",
+	.mode = 0666,
+};
+
+/*
+ * Used only by DMABUF, returns a valid struct device to use as a dummy struct
+ * device for attachment.
+ */
+struct device *iommufd_global_device(void)
+{
+	return iommu_misc_dev.this_device;
+}
+
+static int __init iommufd_init(void)
+{
+	int ret;
+
+	ret = misc_register(&iommu_misc_dev);
+	if (ret)
+		return ret;
+
+	if (IS_ENABLED(CONFIG_IOMMUFD_VFIO_CONTAINER)) {
+		ret = misc_register(&vfio_misc_dev);
+		if (ret)
+			goto err_misc;
+	}
+	ret = iommufd_test_init();
+	if (ret)
+		goto err_vfio_misc;
+	return 0;
+
+err_vfio_misc:
+	if (IS_ENABLED(CONFIG_IOMMUFD_VFIO_CONTAINER))
+		misc_deregister(&vfio_misc_dev);
+err_misc:
+	misc_deregister(&iommu_misc_dev);
+	return ret;
+}
+
+static void __exit iommufd_exit(void)
+{
+	iommufd_test_exit();
+	if (IS_ENABLED(CONFIG_IOMMUFD_VFIO_CONTAINER))
+		misc_deregister(&vfio_misc_dev);
+	misc_deregister(&iommu_misc_dev);
+}
+
+module_init(iommufd_init);
+module_exit(iommufd_exit);
+
+#if IS_ENABLED(CONFIG_IOMMUFD_VFIO_CONTAINER)
+MODULE_ALIAS_MISCDEV(VFIO_MINOR);
+MODULE_ALIAS("devname:vfio/vfio");
+#endif
+MODULE_IMPORT_NS("IOMMUFD_INTERNAL");
+MODULE_IMPORT_NS("IOMMUFD");
+MODULE_IMPORT_NS("DMA_BUF");
+MODULE_DESCRIPTION("I/O Address Space Management for passthrough devices");
+MODULE_LICENSE("GPL");

linux-7.2.3/drivers/iommu/iommufd/vfio_compat.c

@@ -0,0 +1,536 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES
+ */
+#include <linux/file.h>
+#include <linux/interval_tree.h>
+#include <linux/iommu.h>
+#include <linux/iommufd.h>
+#include <linux/slab.h>
+#include <linux/vfio.h>
+#include <uapi/linux/vfio.h>
+#include <uapi/linux/iommufd.h>
+
+#include "iommufd_private.h"
+
+static struct iommufd_ioas *get_compat_ioas(struct iommufd_ctx *ictx)
+{
+	struct iommufd_ioas *ioas = ERR_PTR(-ENODEV);
+
+	xa_lock(&ictx->objects);
+	if (!ictx->vfio_ioas || !iommufd_lock_obj(&ictx->vfio_ioas->obj))
+		goto out_unlock;
+	ioas = ictx->vfio_ioas;
+out_unlock:
+	xa_unlock(&ictx->objects);
+	return ioas;
+}
+
+/**
+ * iommufd_vfio_compat_ioas_get_id - Ensure a compat IOAS exists
+ * @ictx: Context to operate on
+ * @out_ioas_id: The IOAS ID of the compatibility IOAS
+ *
+ * Return the ID of the current compatibility IOAS. The ID can be passed into
+ * other functions that take an ioas_id.
+ */
+int iommufd_vfio_compat_ioas_get_id(struct iommufd_ctx *ictx, u32 *out_ioas_id)
+{
+	struct iommufd_ioas *ioas;
+
+	ioas = get_compat_ioas(ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+	*out_ioas_id = ioas->obj.id;
+	iommufd_put_object(ictx, &ioas->obj);
+	return 0;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_vfio_compat_ioas_get_id, "IOMMUFD_VFIO");
+
+/**
+ * iommufd_vfio_compat_set_no_iommu - Called when a no-iommu device is attached
+ * @ictx: Context to operate on
+ *
+ * This allows selecting the VFIO_NOIOMMU_IOMMU and blocks normal types.
+ */
+int iommufd_vfio_compat_set_no_iommu(struct iommufd_ctx *ictx)
+{
+	int ret;
+
+	xa_lock(&ictx->objects);
+	if (!ictx->vfio_ioas) {
+		ictx->no_iommu_mode = 1;
+		ret = 0;
+	} else {
+		ret = -EINVAL;
+	}
+	xa_unlock(&ictx->objects);
+	return ret;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_vfio_compat_set_no_iommu, "IOMMUFD_VFIO");
+
+/**
+ * iommufd_vfio_compat_ioas_create - Ensure the compat IOAS is created
+ * @ictx: Context to operate on
+ *
+ * The compatibility IOAS is the IOAS that the vfio compatibility ioctls operate
+ * on since they do not have an IOAS ID input in their ABI. Only attaching a
+ * group should cause a default creation of the internal ioas, this does nothing
+ * if an existing ioas has already been assigned somehow.
+ */
+int iommufd_vfio_compat_ioas_create(struct iommufd_ctx *ictx)
+{
+	struct iommufd_ioas *ioas = NULL;
+	int ret;
+
+	ioas = iommufd_ioas_alloc(ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	xa_lock(&ictx->objects);
+	/*
+	 * VFIO won't allow attaching a container to both iommu and no iommu
+	 * operation
+	 */
+	if (ictx->no_iommu_mode) {
+		ret = -EINVAL;
+		goto out_abort;
+	}
+
+	if (ictx->vfio_ioas && iommufd_lock_obj(&ictx->vfio_ioas->obj)) {
+		ret = 0;
+		iommufd_put_object(ictx, &ictx->vfio_ioas->obj);
+		goto out_abort;
+	}
+	ictx->vfio_ioas = ioas;
+	xa_unlock(&ictx->objects);
+
+	/*
+	 * An automatically created compat IOAS is treated as a userspace
+	 * created object. Userspace can learn the ID via IOMMU_VFIO_IOAS_GET,
+	 * and if not manually destroyed it will be destroyed automatically
+	 * at iommufd release.
+	 */
+	iommufd_object_finalize(ictx, &ioas->obj);
+	return 0;
+
+out_abort:
+	xa_unlock(&ictx->objects);
+	iommufd_object_abort(ictx, &ioas->obj);
+	return ret;
+}
+EXPORT_SYMBOL_NS_GPL(iommufd_vfio_compat_ioas_create, "IOMMUFD_VFIO");
+
+int iommufd_vfio_ioas(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_vfio_ioas *cmd = ucmd->cmd;
+	struct iommufd_ioas *ioas;
+
+	if (cmd->__reserved)
+		return -EOPNOTSUPP;
+	switch (cmd->op) {
+	case IOMMU_VFIO_IOAS_GET:
+		ioas = get_compat_ioas(ucmd->ictx);
+		if (IS_ERR(ioas))
+			return PTR_ERR(ioas);
+		cmd->ioas_id = ioas->obj.id;
+		iommufd_put_object(ucmd->ictx, &ioas->obj);
+		return iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+
+	case IOMMU_VFIO_IOAS_SET:
+		ioas = iommufd_get_ioas(ucmd->ictx, cmd->ioas_id);
+		if (IS_ERR(ioas))
+			return PTR_ERR(ioas);
+		xa_lock(&ucmd->ictx->objects);
+		ucmd->ictx->vfio_ioas = ioas;
+		xa_unlock(&ucmd->ictx->objects);
+		iommufd_put_object(ucmd->ictx, &ioas->obj);
+		return 0;
+
+	case IOMMU_VFIO_IOAS_CLEAR:
+		xa_lock(&ucmd->ictx->objects);
+		ucmd->ictx->vfio_ioas = NULL;
+		xa_unlock(&ucmd->ictx->objects);
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int iommufd_vfio_map_dma(struct iommufd_ctx *ictx, unsigned int cmd,
+				void __user *arg)
+{
+	u32 supported_flags = VFIO_DMA_MAP_FLAG_READ | VFIO_DMA_MAP_FLAG_WRITE;
+	size_t minsz = offsetofend(struct vfio_iommu_type1_dma_map, size);
+	struct vfio_iommu_type1_dma_map map;
+	int iommu_prot = IOMMU_CACHE;
+	struct iommufd_ioas *ioas;
+	unsigned long iova;
+	int rc;
+
+	if (copy_from_user(&map, arg, minsz))
+		return -EFAULT;
+
+	if (map.argsz < minsz || map.flags & ~supported_flags)
+		return -EINVAL;
+
+	if (map.flags & VFIO_DMA_MAP_FLAG_READ)
+		iommu_prot |= IOMMU_READ;
+	if (map.flags & VFIO_DMA_MAP_FLAG_WRITE)
+		iommu_prot |= IOMMU_WRITE;
+
+	ioas = get_compat_ioas(ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	/*
+	 * Maps created through the legacy interface always use VFIO compatible
+	 * rlimit accounting. If the user wishes to use the faster user based
+	 * rlimit accounting then they must use the new interface.
+	 */
+	iova = map.iova;
+	rc = iopt_map_user_pages(ictx, &ioas->iopt, &iova, u64_to_user_ptr(map.vaddr),
+				 map.size, iommu_prot, 0);
+	iommufd_put_object(ictx, &ioas->obj);
+	return rc;
+}
+
+static int iommufd_vfio_unmap_dma(struct iommufd_ctx *ictx, unsigned int cmd,
+				  void __user *arg)
+{
+	size_t minsz = offsetofend(struct vfio_iommu_type1_dma_unmap, size);
+	/*
+	 * VFIO_DMA_UNMAP_FLAG_GET_DIRTY_BITMAP is obsoleted by the new
+	 * dirty tracking direction:
+	 *  https://lore.kernel.org/kvm/20220731125503.142683-1-yishaih@nvidia.com/
+	 *  https://lore.kernel.org/kvm/20220428210933.3583-1-joao.m.martins@oracle.com/
+	 */
+	u32 supported_flags = VFIO_DMA_UNMAP_FLAG_ALL;
+	struct vfio_iommu_type1_dma_unmap unmap;
+	unsigned long unmapped = 0;
+	struct iommufd_ioas *ioas;
+	int rc;
+
+	if (copy_from_user(&unmap, arg, minsz))
+		return -EFAULT;
+
+	if (unmap.argsz < minsz || unmap.flags & ~supported_flags)
+		return -EINVAL;
+
+	ioas = get_compat_ioas(ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	if (unmap.flags & VFIO_DMA_UNMAP_FLAG_ALL) {
+		if (unmap.iova != 0 || unmap.size != 0) {
+			rc = -EINVAL;
+			goto err_put;
+		}
+		rc = iopt_unmap_all(&ioas->iopt, &unmapped);
+	} else {
+		if (READ_ONCE(ioas->iopt.disable_large_pages)) {
+			/*
+			 * Create cuts at the start and last of the requested
+			 * range. If the start IOVA is 0 then it doesn't need to
+			 * be cut.
+			 */
+			unsigned long iovas[] = { unmap.iova + unmap.size - 1,
+						  unmap.iova - 1 };
+
+			rc = iopt_cut_iova(&ioas->iopt, iovas,
+					   unmap.iova ? 2 : 1);
+			if (rc)
+				goto err_put;
+		}
+		rc = iopt_unmap_iova(&ioas->iopt, unmap.iova, unmap.size,
+				     &unmapped);
+	}
+	unmap.size = unmapped;
+	if (copy_to_user(arg, &unmap, minsz))
+		rc = -EFAULT;
+
+err_put:
+	iommufd_put_object(ictx, &ioas->obj);
+	return rc;
+}
+
+static int iommufd_vfio_cc_iommu(struct iommufd_ctx *ictx)
+{
+	struct iommufd_hwpt_paging *hwpt_paging;
+	struct iommufd_ioas *ioas;
+	int rc = 1;
+
+	ioas = get_compat_ioas(ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	mutex_lock(&ioas->mutex);
+	list_for_each_entry(hwpt_paging, &ioas->hwpt_list, hwpt_item) {
+		if (!hwpt_paging->enforce_cache_coherency) {
+			rc = 0;
+			break;
+		}
+	}
+	mutex_unlock(&ioas->mutex);
+
+	iommufd_put_object(ictx, &ioas->obj);
+	return rc;
+}
+
+static int iommufd_vfio_check_extension(struct iommufd_ctx *ictx,
+					unsigned long type)
+{
+	switch (type) {
+	case VFIO_TYPE1_IOMMU:
+	case VFIO_TYPE1v2_IOMMU:
+	case VFIO_UNMAP_ALL:
+		return !ictx->no_iommu_mode;
+
+	case VFIO_NOIOMMU_IOMMU:
+		return IS_ENABLED(CONFIG_VFIO_NOIOMMU);
+
+	case VFIO_DMA_CC_IOMMU:
+		return iommufd_vfio_cc_iommu(ictx);
+
+	case __VFIO_RESERVED_TYPE1_NESTING_IOMMU:
+		return 0;
+
+	/*
+	 * VFIO_DMA_MAP_FLAG_VADDR
+	 * https://lore.kernel.org/kvm/1611939252-7240-1-git-send-email-steven.sistare@oracle.com/
+	 * https://lore.kernel.org/all/Yz777bJZjTyLrHEQ@nvidia.com/
+	 *
+	 * It is hard to see how this could be implemented safely.
+	 */
+	case VFIO_UPDATE_VADDR:
+	default:
+		return 0;
+	}
+}
+
+static int iommufd_vfio_set_iommu(struct iommufd_ctx *ictx, unsigned long type)
+{
+	bool no_iommu_mode = READ_ONCE(ictx->no_iommu_mode);
+	struct iommufd_ioas *ioas = NULL;
+	int rc = 0;
+
+	/*
+	 * Emulation for NOIOMMU is imperfect in that VFIO blocks almost all
+	 * other ioctls. We let them keep working but they mostly fail since no
+	 * IOAS should exist.
+	 */
+	if (IS_ENABLED(CONFIG_VFIO_NOIOMMU) && type == VFIO_NOIOMMU_IOMMU &&
+	    no_iommu_mode) {
+		if (!capable(CAP_SYS_RAWIO))
+			return -EPERM;
+		return 0;
+	}
+
+	if ((type != VFIO_TYPE1_IOMMU && type != VFIO_TYPE1v2_IOMMU) ||
+	    no_iommu_mode)
+		return -EINVAL;
+
+	/* VFIO fails the set_iommu if there is no group */
+	ioas = get_compat_ioas(ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	/*
+	 * The difference between TYPE1 and TYPE1v2 is the ability to unmap in
+	 * the middle of mapped ranges. This is complicated by huge page support
+	 * which creates single large IOPTEs that cannot be split by the iommu
+	 * driver. TYPE1 is very old at this point and likely nothing uses it,
+	 * however it is simple enough to emulate by simply disabling the
+	 * problematic large IOPTEs. Then we can safely unmap within any range.
+	 */
+	if (type == VFIO_TYPE1_IOMMU)
+		rc = iopt_disable_large_pages(&ioas->iopt);
+	iommufd_put_object(ictx, &ioas->obj);
+	return rc;
+}
+
+static unsigned long iommufd_get_pagesizes(struct iommufd_ioas *ioas)
+{
+	struct io_pagetable *iopt = &ioas->iopt;
+	unsigned long pgsize_bitmap = ULONG_MAX;
+	struct iommu_domain *domain;
+	unsigned long index;
+
+	down_read(&iopt->domains_rwsem);
+	xa_for_each(&iopt->domains, index, domain)
+		pgsize_bitmap &= domain->pgsize_bitmap;
+
+	/* See vfio_update_pgsize_bitmap() */
+	if (pgsize_bitmap & ~PAGE_MASK) {
+		pgsize_bitmap &= PAGE_MASK;
+		pgsize_bitmap |= PAGE_SIZE;
+	}
+	pgsize_bitmap = max(pgsize_bitmap, ioas->iopt.iova_alignment);
+	up_read(&iopt->domains_rwsem);
+	return pgsize_bitmap;
+}
+
+static int iommufd_fill_cap_iova(struct iommufd_ioas *ioas,
+				 struct vfio_info_cap_header __user *cur,
+				 size_t avail)
+{
+	struct vfio_iommu_type1_info_cap_iova_range __user *ucap_iovas =
+		container_of(cur,
+			     struct vfio_iommu_type1_info_cap_iova_range __user,
+			     header);
+	struct vfio_iommu_type1_info_cap_iova_range cap_iovas = {
+		.header = {
+			.id = VFIO_IOMMU_TYPE1_INFO_CAP_IOVA_RANGE,
+			.version = 1,
+		},
+	};
+	struct interval_tree_span_iter span;
+
+	interval_tree_for_each_span(&span, &ioas->iopt.reserved_itree, 0,
+				    ULONG_MAX) {
+		struct vfio_iova_range range;
+
+		if (!span.is_hole)
+			continue;
+		range.start = span.start_hole;
+		range.end = span.last_hole;
+		if (avail >= struct_size(&cap_iovas, iova_ranges,
+					 cap_iovas.nr_iovas + 1) &&
+		    copy_to_user(&ucap_iovas->iova_ranges[cap_iovas.nr_iovas],
+				 &range, sizeof(range)))
+			return -EFAULT;
+		cap_iovas.nr_iovas++;
+	}
+	if (avail >= struct_size(&cap_iovas, iova_ranges, cap_iovas.nr_iovas) &&
+	    copy_to_user(ucap_iovas, &cap_iovas, sizeof(cap_iovas)))
+		return -EFAULT;
+	return struct_size(&cap_iovas, iova_ranges, cap_iovas.nr_iovas);
+}
+
+static int iommufd_fill_cap_dma_avail(struct iommufd_ioas *ioas,
+				      struct vfio_info_cap_header __user *cur,
+				      size_t avail)
+{
+	struct vfio_iommu_type1_info_dma_avail cap_dma = {
+		.header = {
+			.id = VFIO_IOMMU_TYPE1_INFO_DMA_AVAIL,
+			.version = 1,
+		},
+		/*
+		 * iommufd's limit is based on the cgroup's memory limit.
+		 * Normally vfio would return U16_MAX here, and provide a module
+		 * parameter to adjust it. Since S390 qemu userspace actually
+		 * pays attention and needs a value bigger than U16_MAX return
+		 * U32_MAX.
+		 */
+		.avail = U32_MAX,
+	};
+
+	if (avail >= sizeof(cap_dma) &&
+	    copy_to_user(cur, &cap_dma, sizeof(cap_dma)))
+		return -EFAULT;
+	return sizeof(cap_dma);
+}
+
+static int iommufd_vfio_iommu_get_info(struct iommufd_ctx *ictx,
+				       void __user *arg)
+{
+	typedef int (*fill_cap_fn)(struct iommufd_ioas *ioas,
+				   struct vfio_info_cap_header __user *cur,
+				   size_t avail);
+	static const fill_cap_fn fill_fns[] = {
+		iommufd_fill_cap_dma_avail,
+		iommufd_fill_cap_iova,
+	};
+	size_t minsz = offsetofend(struct vfio_iommu_type1_info, iova_pgsizes);
+	struct vfio_info_cap_header __user *last_cap = NULL;
+	struct vfio_iommu_type1_info info = {};
+	struct iommufd_ioas *ioas;
+	size_t total_cap_size;
+	int rc;
+	int i;
+
+	if (copy_from_user(&info, arg, minsz))
+		return -EFAULT;
+
+	if (info.argsz < minsz)
+		return -EINVAL;
+	minsz = min_t(size_t, info.argsz, sizeof(info));
+
+	ioas = get_compat_ioas(ictx);
+	if (IS_ERR(ioas))
+		return PTR_ERR(ioas);
+
+	info.flags = VFIO_IOMMU_INFO_PGSIZES;
+	info.iova_pgsizes = iommufd_get_pagesizes(ioas);
+	info.cap_offset = 0;
+
+	down_read(&ioas->iopt.iova_rwsem);
+	total_cap_size = sizeof(info);
+	for (i = 0; i != ARRAY_SIZE(fill_fns); i++) {
+		int cap_size;
+
+		if (info.argsz > total_cap_size)
+			cap_size = fill_fns[i](ioas, arg + total_cap_size,
+					       info.argsz - total_cap_size);
+		else
+			cap_size = fill_fns[i](ioas, NULL, 0);
+		if (cap_size < 0) {
+			rc = cap_size;
+			goto out_put;
+		}
+		cap_size = ALIGN(cap_size, sizeof(u64));
+
+		if (last_cap && info.argsz >= total_cap_size &&
+		    put_user(total_cap_size, &last_cap->next)) {
+			rc = -EFAULT;
+			goto out_put;
+		}
+		last_cap = arg + total_cap_size;
+		total_cap_size += cap_size;
+	}
+
+	/*
+	 * If the user did not provide enough space then only some caps are
+	 * returned and the argsz will be updated to the correct amount to get
+	 * all caps.
+	 */
+	if (info.argsz >= total_cap_size)
+		info.cap_offset = sizeof(info);
+	info.argsz = total_cap_size;
+	info.flags |= VFIO_IOMMU_INFO_CAPS;
+	if (copy_to_user(arg, &info, minsz)) {
+		rc = -EFAULT;
+		goto out_put;
+	}
+	rc = 0;
+
+out_put:
+	up_read(&ioas->iopt.iova_rwsem);
+	iommufd_put_object(ictx, &ioas->obj);
+	return rc;
+}
+
+int iommufd_vfio_ioctl(struct iommufd_ctx *ictx, unsigned int cmd,
+		       unsigned long arg)
+{
+	void __user *uarg = (void __user *)arg;
+
+	switch (cmd) {
+	case VFIO_GET_API_VERSION:
+		return VFIO_API_VERSION;
+	case VFIO_SET_IOMMU:
+		return iommufd_vfio_set_iommu(ictx, arg);
+	case VFIO_CHECK_EXTENSION:
+		return iommufd_vfio_check_extension(ictx, arg);
+	case VFIO_IOMMU_GET_INFO:
+		return iommufd_vfio_iommu_get_info(ictx, uarg);
+	case VFIO_IOMMU_MAP_DMA:
+		return iommufd_vfio_map_dma(ictx, cmd, uarg);
+	case VFIO_IOMMU_UNMAP_DMA:
+		return iommufd_vfio_unmap_dma(ictx, cmd, uarg);
+	case VFIO_IOMMU_DIRTY_PAGES:
+	default:
+		return -ENOIOCTLCMD;
+	}
+	return -ENOIOCTLCMD;
+}

linux-7.2.3/drivers/iommu/iommufd/viommu.c

@@ -0,0 +1,440 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2024, NVIDIA CORPORATION & AFFILIATES
+ */
+#include "iommufd_private.h"
+
+void iommufd_viommu_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_viommu *viommu =
+		container_of(obj, struct iommufd_viommu, obj);
+
+	if (viommu->ops && viommu->ops->destroy)
+		viommu->ops->destroy(viommu);
+	refcount_dec(&viommu->hwpt->common.obj.users);
+	xa_destroy(&viommu->vdevs);
+}
+
+int iommufd_viommu_alloc_ioctl(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_viommu_alloc *cmd = ucmd->cmd;
+	const struct iommu_user_data user_data = {
+		.type = cmd->type,
+		.uptr = u64_to_user_ptr(cmd->data_uptr),
+		.len = cmd->data_len,
+	};
+	struct iommufd_hwpt_paging *hwpt_paging;
+	struct iommufd_viommu *viommu;
+	struct iommufd_device *idev;
+	const struct iommu_ops *ops;
+	size_t viommu_size;
+	int rc;
+
+	if (cmd->flags || cmd->type == IOMMU_VIOMMU_TYPE_DEFAULT)
+		return -EOPNOTSUPP;
+
+	idev = iommufd_get_device(ucmd, cmd->dev_id);
+	if (IS_ERR(idev))
+		return PTR_ERR(idev);
+
+	ops = dev_iommu_ops(idev->dev);
+	if (!ops->get_viommu_size || !ops->viommu_init) {
+		rc = -EOPNOTSUPP;
+		goto out_put_idev;
+	}
+
+	viommu_size = ops->get_viommu_size(idev->dev, cmd->type);
+	if (!viommu_size) {
+		rc = -EOPNOTSUPP;
+		goto out_put_idev;
+	}
+
+	/*
+	 * It is a driver bug for providing a viommu_size smaller than the core
+	 * vIOMMU structure size
+	 */
+	if (WARN_ON_ONCE(viommu_size < sizeof(*viommu))) {
+		rc = -EOPNOTSUPP;
+		goto out_put_idev;
+	}
+
+	hwpt_paging = iommufd_get_hwpt_paging(ucmd, cmd->hwpt_id);
+	if (IS_ERR(hwpt_paging)) {
+		rc = PTR_ERR(hwpt_paging);
+		goto out_put_idev;
+	}
+
+	if (!hwpt_paging->nest_parent) {
+		rc = -EINVAL;
+		goto out_put_hwpt;
+	}
+
+	viommu = (struct iommufd_viommu *)_iommufd_object_alloc_ucmd(
+		ucmd, viommu_size, IOMMUFD_OBJ_VIOMMU);
+	if (IS_ERR(viommu)) {
+		rc = PTR_ERR(viommu);
+		goto out_put_hwpt;
+	}
+
+	xa_init(&viommu->vdevs);
+	viommu->type = cmd->type;
+	viommu->ictx = ucmd->ictx;
+	viommu->hwpt = hwpt_paging;
+	refcount_inc(&viommu->hwpt->common.obj.users);
+	INIT_LIST_HEAD(&viommu->veventqs);
+	init_rwsem(&viommu->veventqs_rwsem);
+	/*
+	 * It is the most likely case that a physical IOMMU is unpluggable. A
+	 * pluggable IOMMU instance (if exists) is responsible for refcounting
+	 * on its own.
+	 */
+	viommu->iommu_dev = __iommu_get_iommu_dev(idev->dev);
+
+	rc = ops->viommu_init(viommu, hwpt_paging->common.domain,
+			      user_data.len ? &user_data : NULL);
+	if (rc)
+		goto out_put_hwpt;
+
+	/* It is a driver bug that viommu->ops isn't filled */
+	if (WARN_ON_ONCE(!viommu->ops)) {
+		rc = -EOPNOTSUPP;
+		goto out_put_hwpt;
+	}
+
+	cmd->out_viommu_id = viommu->obj.id;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+
+out_put_hwpt:
+	iommufd_put_object(ucmd->ictx, &hwpt_paging->common.obj);
+out_put_idev:
+	iommufd_put_object(ucmd->ictx, &idev->obj);
+	return rc;
+}
+
+void iommufd_vdevice_abort(struct iommufd_object *obj)
+{
+	struct iommufd_vdevice *vdev =
+		container_of(obj, struct iommufd_vdevice, obj);
+	struct iommufd_viommu *viommu = vdev->viommu;
+	struct iommufd_device *idev = vdev->idev;
+
+	lockdep_assert_held(&idev->igroup->lock);
+
+	if (vdev->destroy)
+		vdev->destroy(vdev);
+	/* xa_cmpxchg is okay to fail if alloc failed xa_cmpxchg previously */
+	xa_cmpxchg(&viommu->vdevs, vdev->virt_id, vdev, NULL, GFP_KERNEL);
+	refcount_dec(&viommu->obj.users);
+	idev->vdev = NULL;
+}
+
+void iommufd_vdevice_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_vdevice *vdev =
+		container_of(obj, struct iommufd_vdevice, obj);
+	struct iommufd_device *idev = vdev->idev;
+	struct iommufd_ctx *ictx = idev->ictx;
+
+	mutex_lock(&idev->igroup->lock);
+	iommufd_vdevice_abort(obj);
+	mutex_unlock(&idev->igroup->lock);
+	iommufd_put_object(ictx, &idev->obj);
+}
+
+int iommufd_vdevice_alloc_ioctl(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_vdevice_alloc *cmd = ucmd->cmd;
+	struct iommufd_vdevice *vdev;
+	size_t vdev_size = sizeof(*vdev);
+	struct iommufd_viommu *viommu;
+	struct iommufd_device *idev;
+	u64 virt_id = cmd->virt_id;
+	int rc = 0;
+
+	/* virt_id indexes an xarray */
+	if (virt_id > ULONG_MAX)
+		return -EINVAL;
+
+	viommu = iommufd_get_viommu(ucmd, cmd->viommu_id);
+	if (IS_ERR(viommu))
+		return PTR_ERR(viommu);
+
+	idev = iommufd_get_device(ucmd, cmd->dev_id);
+	if (IS_ERR(idev)) {
+		rc = PTR_ERR(idev);
+		goto out_put_viommu;
+	}
+
+	if (viommu->iommu_dev != __iommu_get_iommu_dev(idev->dev)) {
+		rc = -EINVAL;
+		goto out_put_idev;
+	}
+
+	mutex_lock(&idev->igroup->lock);
+	if (idev->destroying) {
+		rc = -ENOENT;
+		goto out_unlock_igroup;
+	}
+
+	if (idev->vdev) {
+		rc = -EEXIST;
+		goto out_unlock_igroup;
+	}
+
+	if (viommu->ops && viommu->ops->vdevice_size) {
+		/*
+		 * It is a driver bug for:
+		 * - ops->vdevice_size smaller than the core structure size
+		 * - not implementing a pairing ops->vdevice_init op
+		 */
+		if (WARN_ON_ONCE(viommu->ops->vdevice_size < vdev_size ||
+				 !viommu->ops->vdevice_init)) {
+			rc = -EOPNOTSUPP;
+			goto out_unlock_igroup;
+		}
+		vdev_size = viommu->ops->vdevice_size;
+	}
+
+	vdev = (struct iommufd_vdevice *)_iommufd_object_alloc(
+		ucmd->ictx, vdev_size, IOMMUFD_OBJ_VDEVICE);
+	if (IS_ERR(vdev)) {
+		rc = PTR_ERR(vdev);
+		goto out_unlock_igroup;
+	}
+
+	vdev->virt_id = virt_id;
+	vdev->viommu = viommu;
+	refcount_inc(&viommu->obj.users);
+	/*
+	 * A wait_cnt reference is held on the idev so long as we have the
+	 * pointer. iommufd_device_pre_destroy() will revoke it before the
+	 * idev real destruction.
+	 */
+	vdev->idev = idev;
+
+	/*
+	 * iommufd_device_destroy() delays until idev->vdev is NULL before
+	 * freeing the idev, which only happens once the vdev is finished
+	 * destruction.
+	 */
+	idev->vdev = vdev;
+
+	/*
+	 * Reserve the slot with a zero entry (reads back as NULL) until the
+	 * vdevice_init() op accepts the vDEVICE. Only the xa_* helpers hide a
+	 * reserved entry, so never use a raw xas_* iterator on this xarray.
+	 */
+	rc = xa_insert(&viommu->vdevs, virt_id, NULL, GFP_KERNEL);
+	if (rc) {
+		if (rc == -EBUSY)
+			rc = -EEXIST;
+		goto out_abort;
+	}
+
+	if (viommu->ops && viommu->ops->vdevice_init) {
+		rc = viommu->ops->vdevice_init(vdev);
+		if (rc) {
+			xa_release(&viommu->vdevs, virt_id);
+			goto out_abort;
+		}
+	}
+
+	xa_store(&viommu->vdevs, virt_id, vdev, GFP_KERNEL);
+
+	cmd->out_vdevice_id = vdev->obj.id;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+	if (rc)
+		goto out_abort;
+	iommufd_object_finalize(ucmd->ictx, &vdev->obj);
+	goto out_unlock_igroup;
+
+out_abort:
+	iommufd_object_abort_and_destroy(ucmd->ictx, &vdev->obj);
+out_unlock_igroup:
+	mutex_unlock(&idev->igroup->lock);
+out_put_idev:
+	if (rc)
+		iommufd_put_object(ucmd->ictx, &idev->obj);
+out_put_viommu:
+	iommufd_put_object(ucmd->ictx, &viommu->obj);
+	return rc;
+}
+
+static void iommufd_hw_queue_destroy_access(struct iommufd_ctx *ictx,
+					    struct iommufd_access *access,
+					    u64 base_iova, size_t length)
+{
+	u64 aligned_iova = PAGE_ALIGN_DOWN(base_iova);
+	u64 offset = base_iova - aligned_iova;
+
+	iommufd_access_unpin_pages(access, aligned_iova,
+				   PAGE_ALIGN(length + offset));
+	iommufd_access_detach_internal(access);
+	iommufd_access_destroy_internal(ictx, access);
+}
+
+void iommufd_hw_queue_destroy(struct iommufd_object *obj)
+{
+	struct iommufd_hw_queue *hw_queue =
+		container_of(obj, struct iommufd_hw_queue, obj);
+
+	if (hw_queue->destroy)
+		hw_queue->destroy(hw_queue);
+	if (hw_queue->access)
+		iommufd_hw_queue_destroy_access(hw_queue->viommu->ictx,
+						hw_queue->access,
+						hw_queue->base_addr,
+						hw_queue->length);
+	if (hw_queue->viommu)
+		refcount_dec(&hw_queue->viommu->obj.users);
+}
+
+/*
+ * When the HW accesses the guest queue via physical addresses, the underlying
+ * physical pages of the guest queue must be contiguous. Also, for the security
+ * concern that IOMMUFD_CMD_IOAS_UNMAP could potentially remove the mappings of
+ * the guest queue from the nesting parent iopt while the HW is still accessing
+ * the guest queue memory physically, such a HW queue must require an access to
+ * pin the underlying pages and prevent that from happening.
+ */
+static struct iommufd_access *
+iommufd_hw_queue_alloc_phys(struct iommu_hw_queue_alloc *cmd,
+			    struct iommufd_viommu *viommu, phys_addr_t *base_pa)
+{
+	u64 aligned_iova = PAGE_ALIGN_DOWN(cmd->nesting_parent_iova);
+	u64 offset = cmd->nesting_parent_iova - aligned_iova;
+	struct iommufd_access *access;
+	struct page **pages;
+	size_t max_npages;
+	size_t length;
+	size_t i;
+	int rc;
+
+	/* max_npages = DIV_ROUND_UP(offset + cmd->length, PAGE_SIZE) */
+	if (check_add_overflow(offset, cmd->length, &length))
+		return ERR_PTR(-ERANGE);
+	if (check_add_overflow(length, PAGE_SIZE - 1, &length))
+		return ERR_PTR(-ERANGE);
+	max_npages = length / PAGE_SIZE;
+	/* length needs to be page aligned too */
+	length = max_npages * PAGE_SIZE;
+
+	/*
+	 * Use kvcalloc() to avoid memory fragmentation for a large page array.
+	 * Set __GFP_NOWARN to avoid syzkaller blowups
+	 */
+	pages = kvzalloc_objs(*pages, max_npages, GFP_KERNEL | __GFP_NOWARN);
+	if (!pages)
+		return ERR_PTR(-ENOMEM);
+
+	access = iommufd_access_create_internal(viommu->ictx);
+	if (IS_ERR(access)) {
+		rc = PTR_ERR(access);
+		goto out_free;
+	}
+
+	rc = iommufd_access_attach_internal(access, viommu->hwpt->ioas);
+	if (rc)
+		goto out_destroy;
+
+	rc = iommufd_access_pin_pages(access, aligned_iova, length, pages, 0);
+	if (rc)
+		goto out_detach;
+
+	/* Validate if the underlying physical pages are contiguous */
+	for (i = 1; i < max_npages; i++) {
+		if (page_to_pfn(pages[i]) == page_to_pfn(pages[i - 1]) + 1)
+			continue;
+		rc = -EFAULT;
+		goto out_unpin;
+	}
+
+	*base_pa = (page_to_pfn(pages[0]) << PAGE_SHIFT) + offset;
+	kvfree(pages);
+	return access;
+
+out_unpin:
+	iommufd_access_unpin_pages(access, aligned_iova, length);
+out_detach:
+	iommufd_access_detach_internal(access);
+out_destroy:
+	iommufd_access_destroy_internal(viommu->ictx, access);
+out_free:
+	kvfree(pages);
+	return ERR_PTR(rc);
+}
+
+int iommufd_hw_queue_alloc_ioctl(struct iommufd_ucmd *ucmd)
+{
+	struct iommu_hw_queue_alloc *cmd = ucmd->cmd;
+	struct iommufd_hw_queue *hw_queue;
+	struct iommufd_viommu *viommu;
+	struct iommufd_access *access;
+	size_t hw_queue_size;
+	phys_addr_t base_pa;
+	u64 last;
+	int rc;
+
+	if (cmd->flags || cmd->type == IOMMU_HW_QUEUE_TYPE_DEFAULT)
+		return -EOPNOTSUPP;
+	if (!cmd->length)
+		return -EINVAL;
+	if (check_add_overflow(cmd->nesting_parent_iova, cmd->length - 1,
+			       &last))
+		return -EOVERFLOW;
+
+	viommu = iommufd_get_viommu(ucmd, cmd->viommu_id);
+	if (IS_ERR(viommu))
+		return PTR_ERR(viommu);
+
+	if (!viommu->ops || !viommu->ops->get_hw_queue_size ||
+	    !viommu->ops->hw_queue_init_phys) {
+		rc = -EOPNOTSUPP;
+		goto out_put_viommu;
+	}
+
+	hw_queue_size = viommu->ops->get_hw_queue_size(viommu, cmd->type);
+	if (!hw_queue_size) {
+		rc = -EOPNOTSUPP;
+		goto out_put_viommu;
+	}
+
+	/*
+	 * It is a driver bug for providing a hw_queue_size smaller than the
+	 * core HW queue structure size
+	 */
+	if (WARN_ON_ONCE(hw_queue_size < sizeof(*hw_queue))) {
+		rc = -EOPNOTSUPP;
+		goto out_put_viommu;
+	}
+
+	hw_queue = (struct iommufd_hw_queue *)_iommufd_object_alloc_ucmd(
+		ucmd, hw_queue_size, IOMMUFD_OBJ_HW_QUEUE);
+	if (IS_ERR(hw_queue)) {
+		rc = PTR_ERR(hw_queue);
+		goto out_put_viommu;
+	}
+
+	access = iommufd_hw_queue_alloc_phys(cmd, viommu, &base_pa);
+	if (IS_ERR(access)) {
+		rc = PTR_ERR(access);
+		goto out_put_viommu;
+	}
+
+	hw_queue->viommu = viommu;
+	refcount_inc(&viommu->obj.users);
+	hw_queue->access = access;
+	hw_queue->type = cmd->type;
+	hw_queue->length = cmd->length;
+	hw_queue->base_addr = cmd->nesting_parent_iova;
+
+	rc = viommu->ops->hw_queue_init_phys(hw_queue, cmd->index, base_pa);
+	if (rc)
+		goto out_put_viommu;
+
+	cmd->out_hw_queue_id = hw_queue->obj.id;
+	rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+
+out_put_viommu:
+	iommufd_put_object(ucmd->ictx, &viommu->obj);
+	return rc;
+}