iommufd: Allow pt_id to carry viommu_id for IOMMU_HWPT_ALLOC

Now a vIOMMU holds a shareable nesting parent HWPT. So, it can act like
that nesting parent HWPT to allocate a nested HWPT.

Support that in the IOMMU_HWPT_ALLOC ioctl handler, and update its kdoc.

Also, add an iommufd_viommu_alloc_hwpt_nested helper to allocate a nested
HWPT for a vIOMMU object. Since a vIOMMU object holds the parent hwpt's
refcount already, increase the refcount of the vIOMMU only.

Link: https://patch.msgid.link/r/a0f24f32bfada8b448d17587adcaedeeb50a67ed.1730836219.git.nicolinc@nvidia.com
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

drivers/iommu/iommufd/hw_pagetable.c

@@ -57,7 +57,10 @@ void iommufd_hwpt_nested_destroy(struct iommufd_object *obj)
 		container_of(obj, struct iommufd_hwpt_nested, common.obj);
 
 	__iommufd_hwpt_destroy(&hwpt_nested->common);
-	refcount_dec(&hwpt_nested->parent->common.obj.users);
+	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)
@@ -260,6 +263,58 @@ iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx,
 	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 (!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_nested->viommu = viommu;
+	refcount_inc(&viommu->obj.users);
+	hwpt_nested->parent = viommu->hwpt;
+
+	hwpt->domain =
+		viommu->ops->alloc_domain_nested(viommu, flags, user_data);
+	if (IS_ERR(hwpt->domain)) {
+		rc = PTR_ERR(hwpt->domain);
+		hwpt->domain = NULL;
+		goto out_abort;
+	}
+	hwpt->domain->owner = viommu->iommu_dev->ops;
+
+	if (WARN_ON_ONCE(hwpt->domain->type != IOMMU_DOMAIN_NESTED)) {
+		rc = -EINVAL;
+		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;
@@ -316,6 +371,22 @@ int iommufd_hwpt_alloc(struct iommufd_ucmd *ucmd)
 			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;

drivers/iommu/iommufd/iommufd_private.h

@@ -290,6 +290,7 @@ struct iommufd_hwpt_paging {
 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)

include/uapi/linux/iommufd.h

@@ -435,7 +435,7 @@ enum iommu_hwpt_data_type {
  * @size: sizeof(struct iommu_hwpt_alloc)
  * @flags: Combination of enum iommufd_hwpt_alloc_flags
  * @dev_id: The device to allocate this HWPT for
- * @pt_id: The IOAS or HWPT to connect this HWPT to
+ * @pt_id: The IOAS or HWPT or vIOMMU to connect this HWPT to
  * @out_hwpt_id: The ID of the new HWPT
  * @__reserved: Must be 0
  * @data_type: One of enum iommu_hwpt_data_type
@@ -454,11 +454,13 @@ enum iommu_hwpt_data_type {
  * IOMMU_HWPT_DATA_NONE. The HWPT can be allocated as a parent HWPT for a
  * nesting configuration by passing IOMMU_HWPT_ALLOC_NEST_PARENT via @flags.
  *
- * A user-managed nested HWPT will be created from a given parent HWPT via
- * @pt_id, in which the parent HWPT must be allocated previously via the
- * same ioctl from a given IOAS (@pt_id). In this case, the @data_type
- * must be set to a pre-defined type corresponding to an I/O page table
- * type supported by the underlying IOMMU hardware.
+ * A user-managed nested HWPT will be created from a given vIOMMU (wrapping a
+ * parent HWPT) or a parent HWPT via @pt_id, in which the parent HWPT must be
+ * allocated previously via the same ioctl from a given IOAS (@pt_id). In this
+ * case, the @data_type must be set to a pre-defined type corresponding to an
+ * I/O page table type supported by the underlying IOMMU hardware. The device
+ * via @dev_id and the vIOMMU via @pt_id must be associated to the same IOMMU
+ * instance.
  *
  * If the @data_type is set to IOMMU_HWPT_DATA_NONE, @data_len and
  * @data_uptr should be zero. Otherwise, both @data_len and @data_uptr