From: Pranjal Shrivastava <praan@google.com>
Date: Fri, 16 Aug 2024 10:49:06 +0000
Subject: iommu: Handle iommu faults for a bad iopf setup
Git-commit: b58b133e680b20d219940e0fdb6f6132c2b60f38
Patch-mainline: v6.11-rc6
References: jsc#PED-10539

The iommu_report_device_fault function was updated to return void while
assuming that drivers only need to call iommu_report_device_fault() for
reporting an iopf. This implementation causes following problems:

1. The drivers rely on the core code to call it's page_reponse,
   however, when a fault is received and no fault capable domain is
   attached / iopf_param is NULL, the ops->page_response is NOT called
   causing the device to stall in case the fault type was PAGE_REQ.

2. The arm_smmu_v3 driver relies on the returned value to log errors
   returning void from iommu_report_device_fault causes these events to
   be missed while logging.

Modify the iommu_report_device_fault function to return -EINVAL for
cases where no fault capable domain is attached or iopf_param was NULL
and calls back to the driver (ops->page_response) in case the fault type
was IOMMU_FAULT_PAGE_REQ. The returned value can be used by the drivers
to log the fault/event as needed.

Reported-by: Kunkun Jiang <jiangkunkun@huawei.com>
Closes: https://lore.kernel.org/all/6147caf0-b9a0-30ca-795e-a1aa502a5c51@huawei.com/
Fixes: 3dfa64aecbaf ("iommu: Make iommu_report_device_fault() return void")
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Pranjal Shrivastava <praan@google.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20240816104906.1010626-1-praan@google.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c |    2 
 drivers/iommu/io-pgfault.c                  |  121 +++++++++++++++++++---------
 include/linux/iommu.h                       |    5 -
 3 files changed, 87 insertions(+), 41 deletions(-)

--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -1777,7 +1777,7 @@ static int arm_smmu_handle_evt(struct ar
 		goto out_unlock;
 	}
 
-	iommu_report_device_fault(master->dev, &fault_evt);
+	ret = iommu_report_device_fault(master->dev, &fault_evt);
 out_unlock:
 	mutex_unlock(&smmu->streams_mutex);
 	return ret;
--- a/drivers/iommu/io-pgfault.c
+++ b/drivers/iommu/io-pgfault.c
@@ -115,6 +115,59 @@ static struct iopf_group *iopf_group_all
 	return group;
 }
 
+static struct iommu_attach_handle *find_fault_handler(struct device *dev,
+						     struct iopf_fault *evt)
+{
+	struct iommu_fault *fault = &evt->fault;
+	struct iommu_attach_handle *attach_handle;
+
+	if (fault->prm.flags & IOMMU_FAULT_PAGE_REQUEST_PASID_VALID) {
+		attach_handle = iommu_attach_handle_get(dev->iommu_group,
+				fault->prm.pasid, 0);
+		if (IS_ERR(attach_handle)) {
+			const struct iommu_ops *ops = dev_iommu_ops(dev);
+
+			if (!ops->user_pasid_table)
+				return NULL;
+			/*
+			 * The iommu driver for this device supports user-
+			 * managed PASID table. Therefore page faults for
+			 * any PASID should go through the NESTING domain
+			 * attached to the device RID.
+			 */
+			attach_handle = iommu_attach_handle_get(
+					dev->iommu_group, IOMMU_NO_PASID,
+					IOMMU_DOMAIN_NESTED);
+			if (IS_ERR(attach_handle))
+				return NULL;
+		}
+	} else {
+		attach_handle = iommu_attach_handle_get(dev->iommu_group,
+				IOMMU_NO_PASID, 0);
+
+		if (IS_ERR(attach_handle))
+			return NULL;
+	}
+
+	if (!attach_handle->domain->iopf_handler)
+		return NULL;
+
+	return attach_handle;
+}
+
+static void iopf_error_response(struct device *dev, struct iopf_fault *evt)
+{
+	const struct iommu_ops *ops = dev_iommu_ops(dev);
+	struct iommu_fault *fault = &evt->fault;
+	struct iommu_page_response resp = {
+		.pasid = fault->prm.pasid,
+		.grpid = fault->prm.grpid,
+		.code = IOMMU_PAGE_RESP_INVALID
+	};
+
+	ops->page_response(dev, evt, &resp);
+}
+
 /**
  * iommu_report_device_fault() - Report fault event to device driver
  * @dev: the device
@@ -153,24 +206,39 @@ static struct iopf_group *iopf_group_all
  * handling framework should guarantee that the iommu domain could only be
  * freed after the device has stopped generating page faults (or the iommu
  * hardware has been set to block the page faults) and the pending page faults
- * have been flushed.
+ * have been flushed. In case no page fault handler is attached or no iopf params
+ * are setup, then the ops->page_response() is called to complete the evt.
+ *
+ * Returns 0 on success, or an error in case of a bad/failed iopf setup.
  */
-void iommu_report_device_fault(struct device *dev, struct iopf_fault *evt)
+int iommu_report_device_fault(struct device *dev, struct iopf_fault *evt)
 {
+	struct iommu_attach_handle *attach_handle;
 	struct iommu_fault *fault = &evt->fault;
 	struct iommu_fault_param *iopf_param;
 	struct iopf_group abort_group = {};
 	struct iopf_group *group;
 
+	attach_handle = find_fault_handler(dev, evt);
+	if (!attach_handle)
+		goto err_bad_iopf;
+
+	/*
+	 * Something has gone wrong if a fault capable domain is attached but no
+	 * iopf_param is setup
+	 */
 	iopf_param = iopf_get_dev_fault_param(dev);
 	if (WARN_ON(!iopf_param))
-		return;
+		goto err_bad_iopf;
 
 	if (!(fault->prm.flags & IOMMU_FAULT_PAGE_REQUEST_LAST_PAGE)) {
-		report_partial_fault(iopf_param, fault);
+		int ret;
+
+		ret = report_partial_fault(iopf_param, fault);
 		iopf_put_dev_fault_param(iopf_param);
 		/* A request that is not the last does not need to be ack'd */
-		return;
+
+		return ret;
 	}
 
 	/*
@@ -185,38 +253,7 @@ void iommu_report_device_fault(struct de
 	if (group == &abort_group)
 		goto err_abort;
 
-	if (fault->prm.flags & IOMMU_FAULT_PAGE_REQUEST_PASID_VALID) {
-		group->attach_handle = iommu_attach_handle_get(dev->iommu_group,
-							       fault->prm.pasid,
-							       0);
-		if (IS_ERR(group->attach_handle)) {
-			const struct iommu_ops *ops = dev_iommu_ops(dev);
-
-			if (!ops->user_pasid_table)
-				goto err_abort;
-
-			/*
-			 * The iommu driver for this device supports user-
-			 * managed PASID table. Therefore page faults for
-			 * any PASID should go through the NESTING domain
-			 * attached to the device RID.
-			 */
-			group->attach_handle =
-				iommu_attach_handle_get(dev->iommu_group,
-							IOMMU_NO_PASID,
-							IOMMU_DOMAIN_NESTED);
-			if (IS_ERR(group->attach_handle))
-				goto err_abort;
-		}
-	} else {
-		group->attach_handle =
-			iommu_attach_handle_get(dev->iommu_group, IOMMU_NO_PASID, 0);
-		if (IS_ERR(group->attach_handle))
-			goto err_abort;
-	}
-
-	if (!group->attach_handle->domain->iopf_handler)
-		goto err_abort;
+	group->attach_handle = attach_handle;
 
 	/*
 	 * On success iopf_handler must call iopf_group_response() and
@@ -225,7 +262,7 @@ void iommu_report_device_fault(struct de
 	if (group->attach_handle->domain->iopf_handler(group))
 		goto err_abort;
 
-	return;
+	return 0;
 
 err_abort:
 	dev_warn_ratelimited(dev, "iopf with pasid %d aborted\n",
@@ -235,6 +272,14 @@ err_abort:
 		__iopf_free_group(group);
 	else
 		iopf_free_group(group);
+
+	return 0;
+
+err_bad_iopf:
+	if (fault->type == IOMMU_FAULT_PAGE_REQ)
+		iopf_error_response(dev, evt);
+
+	return -EINVAL;
 }
 EXPORT_SYMBOL_GPL(iommu_report_device_fault);
 
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -1565,7 +1565,7 @@ struct iopf_queue *iopf_queue_alloc(cons
 void iopf_queue_free(struct iopf_queue *queue);
 int iopf_queue_discard_partial(struct iopf_queue *queue);
 void iopf_free_group(struct iopf_group *group);
-void iommu_report_device_fault(struct device *dev, struct iopf_fault *evt);
+int iommu_report_device_fault(struct device *dev, struct iopf_fault *evt);
 void iopf_group_response(struct iopf_group *group,
 			 enum iommu_page_response_code status);
 #else
@@ -1603,9 +1603,10 @@ static inline void iopf_free_group(struc
 {
 }
 
-static inline void
+static inline int
 iommu_report_device_fault(struct device *dev, struct iopf_fault *evt)
 {
+	return -ENODEV;
 }
 
 static inline void iopf_group_response(struct iopf_group *group,
