iommu: Avoid races around device probe

We currently have 3 different ways that __iommu_probe_device() may be
called, but no real guarantee that multiple callers can't tread on each
other, especially once asynchronous driver probe gets involved. It would
likely have taken a fair bit of luck to hit this previously, but commit
57365a04c921 ("iommu: Move bus setup to IOMMU device registration") ups
the odds since now it's not just omap-iommu that may trigger multiple
bus_iommu_probe() calls in parallel if probing asynchronously.

Add a lock to ensure we can't try to double-probe a device, and also
close some possible race windows to make sure we're truly robust against
trying to double-initialise a group via two different member devices.

Reported-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Brian Norris <briannorris@chromium.org>
Fixes: 57365a04c921 ("iommu: Move bus setup to IOMMU device registration")
Link: https://lore.kernel.org/r/1946ef9f774851732eed78760a78ec40dbc6d178.1667591503.git.robin.murphy@arm.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>

1 files changed, 22 insertions, 6 deletions

diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 6ca377f4fbf9..7c99d8eb3182 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -306,13 +306,23 @@ static int __iommu_probe_device(struct device *dev, struct list_head *group_list
 	const struct iommu_ops *ops = dev->bus->iommu_ops;
 	struct iommu_device *iommu_dev;
 	struct iommu_group *group;
+	static DEFINE_MUTEX(iommu_probe_device_lock);
 	int ret;
 
 	if (!ops)
 		return -ENODEV;
-
-	if (!dev_iommu_get(dev))
-		return -ENOMEM;
+	/*
+	 * Serialise to avoid races between IOMMU drivers registering in
+	 * parallel and/or the "replay" calls from ACPI/OF code via client
+	 * driver probe. Once the latter have been cleaned up we should
+	 * probably be able to use device_lock() here to minimise the scope,
+	 * but for now enforcing a simple global ordering is fine.
+	 */
+	mutex_lock(&iommu_probe_device_lock);
+	if (!dev_iommu_get(dev)) {
+		ret = -ENOMEM;
+		goto err_unlock;
+	}
 
 	if (!try_module_get(ops->owner)) {
 		ret = -EINVAL;
@@ -333,11 +343,14 @@ static int __iommu_probe_device(struct device *dev, struct list_head *group_list
 		ret = PTR_ERR(group);
 		goto out_release;
 	}
-	iommu_group_put(group);
 
+	mutex_lock(&group->mutex);
 	if (group_list && !group->default_domain && list_empty(&group->entry))
 		list_add_tail(&group->entry, group_list);
+	mutex_unlock(&group->mutex);
+	iommu_group_put(group);
 
+	mutex_unlock(&iommu_probe_device_lock);
 	iommu_device_link(iommu_dev, dev);
 
 	return 0;
@@ -352,6 +365,9 @@ out_module_put:
 err_free:
 	dev_iommu_free(dev);
 
+err_unlock:
+	mutex_unlock(&iommu_probe_device_lock);
+
 	return ret;
 }
 
@@ -1824,11 +1840,11 @@ int bus_iommu_probe(struct bus_type *bus)
 		return ret;
 
 	list_for_each_entry_safe(group, next, &group_list, entry) {
+		mutex_lock(&group->mutex);
+
 		/* Remove item from the list */
 		list_del_init(&group->entry);
 
-		mutex_lock(&group->mutex);
-
 		/* Try to allocate default domain */
 		probe_alloc_default_domain(bus, group);