From: Robin Murphy <robin.murphy@xxxxxxx>
[ Upstream commit 01657bc14a3990c665375f77978631fee77b1fce ]
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@xxxxxxxxxxxx>
Signed-off-by: Robin Murphy <robin.murphy@xxxxxxx>
Tested-by: Brian Norris <briannorris@xxxxxxxxxxxx>
Fixes: 57365a04c921 ("iommu: Move bus setup to IOMMU device registration")
Link: https://lore.kernel.org/r/1946ef9f774851732eed78760a78ec40dbc6d178.1667591503.git.robin.murphy@xxxxxxx
Signed-off-by: Joerg Roedel <jroedel@xxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/iommu/iommu.c | 28 ++++++++++++++++++++++------
1 file changed, 22 insertions(+), 6 deletions(-)
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 65a3b3d886dc..959d895fc1df 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -283,13 +283,23 @@ 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;
@@ -309,11 +319,14 @@ 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;
@@ -328,6 +341,9 @@ static int __iommu_probe_device(struct device *dev, struct list_head *group_list
err_free:
dev_iommu_free(dev);
+err_unlock:
+ mutex_unlock(&iommu_probe_device_lock);
+
return ret;
}
@@ -1799,11 +1815,11 @@ 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);
--
2.35.1
