iommu/vt-d: Fix PCI device refcount leak in prq_event_thread()
authorYang Yingliang <yangyingliang@huawei.com>
Thu, 1 Dec 2022 04:01:25 +0000 (12:01 +0800)
committerJoerg Roedel <jroedel@suse.de>
Fri, 2 Dec 2022 10:45:32 +0000 (11:45 +0100)
As comment of pci_get_domain_bus_and_slot() says, it returns a pci device
with refcount increment, when finish using it, the caller must decrease
the reference count by calling pci_dev_put(). So call pci_dev_put() after
using the 'pdev' to avoid refcount leak.

Besides, if the 'pdev' is null or intel_svm_prq_report() returns error,
there is no need to trace this fault.

Fixes: 06f4b8d09dba ("iommu/vt-d: Remove unnecessary SVA data accesses in page fault path")
Suggested-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20221119144028.2452731-1-yangyingliang@huawei.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
drivers/iommu/intel/svm.c

index fe615c5..03b2535 100644 (file)
@@ -748,12 +748,16 @@ bad_req:
                 * If prq is to be handled outside iommu driver via receiver of
                 * the fault notifiers, we skip the page response here.
                 */
-               if (!pdev || intel_svm_prq_report(iommu, &pdev->dev, req))
-                       handle_bad_prq_event(iommu, req, QI_RESP_INVALID);
+               if (!pdev)
+                       goto bad_req;
 
-               trace_prq_report(iommu, &pdev->dev, req->qw_0, req->qw_1,
-                                req->priv_data[0], req->priv_data[1],
-                                iommu->prq_seq_number++);
+               if (intel_svm_prq_report(iommu, &pdev->dev, req))
+                       handle_bad_prq_event(iommu, req, QI_RESP_INVALID);
+               else
+                       trace_prq_report(iommu, &pdev->dev, req->qw_0, req->qw_1,
+                                        req->priv_data[0], req->priv_data[1],
+                                        iommu->prq_seq_number++);
+               pci_dev_put(pdev);
 prq_advance:
                head = (head + sizeof(*req)) & PRQ_RING_MASK;
        }