[PATCH] iommufd: Enforce IOMMU_RESV_SW_MSI upon hwpt_paging allocation

[Nicolin Chen]

IOMMU_RESV_SW_MSI is a unique region defined by an IOMMU driver. Though it
is eventually used by a device for address translation to an MSI location
(including nested cases), practically it is a universal region across all
domains allocated for the IOMMU that defines it.

Currently IOMMUFD core fetches and reserves the region during an attach to
an hwpt_paging. It works with a hwpt_paging-only case, but might not work
with a nested case where a device could directly attach to a hwpt_nested,
bypassing the hwpt_paging attachment.

Move the enforcement forward, to the hwpt_paging allocation function. Then
clean up all the SW_MSI related things in the attach/replace routine.

Signed-off-by: Nicolin Chen <nicolinc@xxxxxxxxxx>
---
 drivers/iommu/iommufd/device.c          | 67 ++-----------------------
 drivers/iommu/iommufd/hw_pagetable.c    | 45 +++++++++++++++++
 drivers/iommu/iommufd/io_pagetable.c    | 13 +++--
 drivers/iommu/iommufd/iommufd_private.h |  5 +-
 4 files changed, 55 insertions(+), 75 deletions(-)

diff --git a/drivers/iommu/iommufd/device.c b/drivers/iommu/iommufd/device.c
index 9a7ec5997c61c..bc8baee32a9da 100644
--- a/drivers/iommu/iommufd/device.c
+++ b/drivers/iommu/iommufd/device.c
@@ -90,7 +90,6 @@ static struct iommufd_group *iommufd_get_group(struct iommufd_ctx *ictx,
 	kref_init(&new_igroup->ref);
 	mutex_init(&new_igroup->lock);
 	INIT_LIST_HEAD(&new_igroup->device_list);
-	new_igroup->sw_msi_start = PHYS_ADDR_MAX;
 	/* group reference moves into new_igroup */
 	new_igroup->group = group;
 
@@ -293,64 +292,6 @@ u32 iommufd_device_to_id(struct iommufd_device *idev)
 }
 EXPORT_SYMBOL_NS_GPL(iommufd_device_to_id, IOMMUFD);
 
-static int iommufd_group_setup_msi(struct iommufd_group *igroup,
-				   struct iommufd_hwpt_paging *hwpt_paging)
-{
-	phys_addr_t sw_msi_start = igroup->sw_msi_start;
-	int rc;
-
-	/*
-	 * If the IOMMU driver gives a IOMMU_RESV_SW_MSI then it is asking us to
-	 * call iommu_get_msi_cookie() on its behalf. This is necessary to setup
-	 * the MSI window so iommu_dma_prepare_msi() can install pages into our
-	 * domain after request_irq(). If it is not done interrupts will not
-	 * work on this domain.
-	 *
-	 * FIXME: This is conceptually broken for iommufd since we want to allow
-	 * userspace to change the domains, eg switch from an identity IOAS to a
-	 * DMA IOAS. There is currently no way to create a MSI window that
-	 * matches what the IRQ layer actually expects in a newly created
-	 * domain.
-	 */
-	if (sw_msi_start != PHYS_ADDR_MAX && !hwpt_paging->msi_cookie) {
-		rc = iommu_get_msi_cookie(hwpt_paging->common.domain,
-					  sw_msi_start);
-		if (rc)
-			return rc;
-
-		/*
-		 * iommu_get_msi_cookie() can only be called once per domain,
-		 * it returns -EBUSY on later calls.
-		 */
-		hwpt_paging->msi_cookie = true;
-	}
-	return 0;
-}
-
-static int iommufd_hwpt_paging_attach(struct iommufd_hwpt_paging *hwpt_paging,
-				      struct iommufd_device *idev)
-{
-	int rc;
-
-	lockdep_assert_held(&idev->igroup->lock);
-
-	rc = iopt_table_enforce_dev_resv_regions(&hwpt_paging->ioas->iopt,
-						 idev->dev,
-						 &idev->igroup->sw_msi_start);
-	if (rc)
-		return rc;
-
-	if (list_empty(&idev->igroup->device_list)) {
-		rc = iommufd_group_setup_msi(idev->igroup, hwpt_paging);
-		if (rc) {
-			iopt_remove_reserved_iova(&hwpt_paging->ioas->iopt,
-						  idev->dev);
-			return rc;
-		}
-	}
-	return 0;
-}
-
 int iommufd_hw_pagetable_attach(struct iommufd_hw_pagetable *hwpt,
 				struct iommufd_device *idev)
 {
@@ -364,7 +305,8 @@ int iommufd_hw_pagetable_attach(struct iommufd_hw_pagetable *hwpt,
 	}
 
 	if (hwpt_is_paging(hwpt)) {
-		rc = iommufd_hwpt_paging_attach(to_hwpt_paging(hwpt), idev);
+		rc = iopt_table_enforce_dev_resv_regions(
+				&to_hwpt_paging(hwpt)->ioas->iopt, idev->dev);
 		if (rc)
 			goto err_unlock;
 	}
@@ -453,15 +395,12 @@ iommufd_group_do_replace_paging(struct iommufd_group *igroup,
 	    hwpt_paging->ioas != to_hwpt_paging(old_hwpt)->ioas) {
 		list_for_each_entry(cur, &igroup->device_list, group_item) {
 			rc = iopt_table_enforce_dev_resv_regions(
-				&hwpt_paging->ioas->iopt, cur->dev, NULL);
+				&hwpt_paging->ioas->iopt, cur->dev);
 			if (rc)
 				goto err_unresv;
 		}
 	}
 
-	rc = iommufd_group_setup_msi(igroup, hwpt_paging);
-	if (rc)
-		goto err_unresv;
 	return 0;
 
 err_unresv:
diff --git a/drivers/iommu/iommufd/hw_pagetable.c b/drivers/iommu/iommufd/hw_pagetable.c
index aefde4443671e..dfb132e4dfbd2 100644
--- a/drivers/iommu/iommufd/hw_pagetable.c
+++ b/drivers/iommu/iommufd/hw_pagetable.c
@@ -82,6 +82,42 @@ iommufd_hwpt_paging_enforce_cc(struct iommufd_hwpt_paging *hwpt_paging)
 	return 0;
 }
 
+static int
+iommufd_hwpt_paging_enforce_sw_msi(struct iommufd_hwpt_paging *hwpt_paging,
+				   struct iommufd_device *idev)
+{
+	struct iommu_domain *domain = hwpt_paging->common.domain;
+	struct io_pagetable *iopt = &hwpt_paging->ioas->iopt;
+	phys_addr_t sw_msi_start = PHYS_ADDR_MAX;
+	struct iommu_resv_region *resv;
+	LIST_HEAD(resv_regions);
+	int rc = 0;
+
+	if (iommufd_should_fail())
+		return -EINVAL;
+
+	/* FIXME: drivers allocate memory but there is no failure propogated */
+	iommu_get_resv_regions(idev->dev, &resv_regions);
+	list_for_each_entry(resv, &resv_regions, list) {
+		if (resv->type != IOMMU_RESV_SW_MSI)
+			continue;
+		down_write(&iopt->iova_rwsem);
+		/* owner=domain so that abort/destroy() can clean it up */
+		rc = iopt_reserve_iova(iopt, resv->start,
+				       resv->length - 1 + resv->start, domain);
+		up_write(&iopt->iova_rwsem);
+		if (!rc)
+			sw_msi_start = resv->start;
+		break;
+	}
+	iommu_put_resv_regions(idev->dev, &resv_regions);
+
+	if (sw_msi_start == PHYS_ADDR_MAX)
+		return rc;
+
+	return iommu_get_msi_cookie(domain, sw_msi_start);
+}
+
 /**
  * iommufd_hwpt_paging_alloc() - Get a PAGING iommu_domain for a device
  * @ictx: iommufd context
@@ -173,6 +209,15 @@ iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas,
 			goto out_abort;
 	}
 
+	/*
+	 * IOMMU_RESV_SW_MSI is a universal per-IOMMU IOVA region arbitrarily
+	 * defined by a driver. Any hw_pagetable that is allocated for such an
+	 * IOMMU must enforce the region in its reserved space.
+	 */
+	rc = iommufd_hwpt_paging_enforce_sw_msi(hwpt_paging, idev);
+	if (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
diff --git a/drivers/iommu/iommufd/io_pagetable.c b/drivers/iommu/iommufd/io_pagetable.c
index 05fd9d3abf1b8..c9b7c7f6e046b 100644
--- a/drivers/iommu/iommufd/io_pagetable.c
+++ b/drivers/iommu/iommufd/io_pagetable.c
@@ -1368,8 +1368,7 @@ void iopt_remove_access(struct io_pagetable *iopt,
 
 /* Narrow the valid_iova_itree to include reserved ranges from a device. */
 int iopt_table_enforce_dev_resv_regions(struct io_pagetable *iopt,
-					struct device *dev,
-					phys_addr_t *sw_msi_start)
+					struct device *dev)
 {
 	struct iommu_resv_region *resv;
 	LIST_HEAD(resv_regions);
@@ -1387,14 +1386,14 @@ int iopt_table_enforce_dev_resv_regions(struct io_pagetable *iopt,
 	list_for_each_entry(resv, &resv_regions, list) {
 		if (resv->type == IOMMU_RESV_DIRECT_RELAXABLE)
 			continue;
-
-		if (sw_msi_start && resv->type == IOMMU_RESV_MSI)
-			num_hw_msi++;
-		if (sw_msi_start && resv->type == IOMMU_RESV_SW_MSI) {
-			*sw_msi_start = resv->start;
+		if (resv->type == IOMMU_RESV_SW_MSI) {
 			num_sw_msi++;
+			continue;
 		}
 
+		if (resv->type == IOMMU_RESV_MSI)
+			num_hw_msi++;
+
 		rc = iopt_reserve_iova(iopt, resv->start,
 				       resv->length - 1 + resv->start, dev);
 		if (rc)
diff --git a/drivers/iommu/iommufd/iommufd_private.h b/drivers/iommu/iommufd/iommufd_private.h
index 92efe30a8f0d0..d61ea73776261 100644
--- a/drivers/iommu/iommufd/iommufd_private.h
+++ b/drivers/iommu/iommufd/iommufd_private.h
@@ -89,8 +89,7 @@ int iopt_table_add_domain(struct io_pagetable *iopt,
 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);
+					struct device *dev);
 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,
@@ -302,7 +301,6 @@ struct iommufd_hwpt_paging {
 	struct iommufd_ioas *ioas;
 	bool auto_domain : 1;
 	bool enforce_cache_coherency : 1;
-	bool msi_cookie : 1;
 	bool nest_parent : 1;
 	/* Head at iommufd_ioas::hwpt_list */
 	struct list_head hwpt_item;
@@ -382,7 +380,6 @@ struct iommufd_group {
 	struct iommu_group *group;
 	struct iommufd_hw_pagetable *hwpt;
 	struct list_head device_list;
-	phys_addr_t sw_msi_start;
 };
 
 /*
-- 
2.43.0