diff mbox series

[v6,1/7] iommu: Prevent pasid attach if no ops->remove_dev_pasid

Message ID 20241204122928.11987-2-yi.l.liu@intel.com (mailing list archive)
State New
Headers show
Series Support attaching PASID to the blocked_domain | expand

Commit Message

Yi Liu Dec. 4, 2024, 12:29 p.m. UTC 
driver should implement both set_dev_pasid and remove_dev_pasid op, otherwise
it is a problem how to detach pasid. In reality, it is impossible that an
iommu driver implements set_dev_pasid() but no remove_dev_pasid() op. However,
it is better to check it.

Move the group check to be the first as dev_iommu_ops() may fail when there
is no valid group. Also take the chance to remove the dev_has_iommu() check
as it is duplicated to the group check.

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
---
 drivers/iommu/iommu.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

Comments

Baolu Lu Dec. 5, 2024, 2:13 a.m. UTC | #1 
On 12/4/24 20:29, Yi Liu wrote:
> driver should implement both set_dev_pasid and remove_dev_pasid op, otherwise
> it is a problem how to detach pasid. In reality, it is impossible that an
> iommu driver implements set_dev_pasid() but no remove_dev_pasid() op. However,
> it is better to check it.
> 
> Move the group check to be the first as dev_iommu_ops() may fail when there
> is no valid group. Also take the chance to remove the dev_has_iommu() check
> as it is duplicated to the group check.
> 
> Reviewed-by: Jason Gunthorpe<jgg@nvidia.com>
> Signed-off-by: Yi Liu<yi.l.liu@intel.com>

Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Tian, Kevin Dec. 11, 2024, 8:24 a.m. UTC | #2 
> From: Liu, Yi L <yi.l.liu@intel.com>
> Sent: Wednesday, December 4, 2024 8:29 PM
> 
> driver should implement both set_dev_pasid and remove_dev_pasid op,
> otherwise
> it is a problem how to detach pasid. In reality, it is impossible that an
> iommu driver implements set_dev_pasid() but no remove_dev_pasid() op.
> However,
> it is better to check it.
> 
> Move the group check to be the first as dev_iommu_ops() may fail when
> there
> is no valid group. Also take the chance to remove the dev_has_iommu()
> check
> as it is duplicated to the group check.
> 
> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
> Signed-off-by: Yi Liu <yi.l.liu@intel.com>

Reviewed-by: Kevin Tian <kevin.tian@intel.com>
diff mbox series

Patch

diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 9bc0c74cca3c..38c3f67e441f 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -3368,16 +3368,19 @@  int iommu_attach_device_pasid(struct iommu_domain *domain,
 	/* Caller must be a probed driver on dev */
 	struct iommu_group *group = dev->iommu_group;
 	struct group_device *device;
+	const struct iommu_ops *ops;
 	int ret;
 
-	if (!domain->ops->set_dev_pasid)
-		return -EOPNOTSUPP;
-
 	if (!group)
 		return -ENODEV;
 
-	if (!dev_has_iommu(dev) || dev_iommu_ops(dev) != domain->owner ||
-	    pasid == IOMMU_NO_PASID)
+	ops = dev_iommu_ops(dev);
+
+	if (!domain->ops->set_dev_pasid ||
+	    !ops->remove_dev_pasid)
+		return -EOPNOTSUPP;
+
+	if (ops != domain->owner || pasid == IOMMU_NO_PASID)
 		return -EINVAL;
 
 	mutex_lock(&group->mutex);