iommu/vt-d: Fix IOMMU lookup for SR-IOV Virtual Functions

The VT-d specification (§8.3.3) says:
    ‘Virtual Functions’ of a ‘Physical Function’ are under the scope
    of the same remapping unit as the ‘Physical Function’.

The BIOS is not required to list all the possible VFs in the scope
tables, and arguably *shouldn't* make any attempt to do so, since there
could be a huge number of them.

This has been broken basically for ever — the VF is never going to match
against a specific unit's scope, so it ends up being assigned to the
INCLUDE_ALL IOMMU. Which was always actually correct by coincidence, but
now we're looking at Root-Complex integrated devices with SR-IOV support
it's going to start being wrong.

Fix it to simply use pci_physfn() before doing the lookup for PCI devices.

Cc: stable@vger.kernel.org
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>

2 files changed, 16 insertions, 1 deletions

diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c
index 58470f5ced04..8c53748a769d 100644
--- a/drivers/iommu/dmar.c
+++ b/drivers/iommu/dmar.c
@@ -338,7 +338,9 @@ static int dmar_pci_bus_notifier(struct notifier_block *nb,
         struct pci_dev *pdev = to_pci_dev(data);
         struct dmar_pci_notify_info *info;
 
-        /* Only care about add/remove events for physical functions */
+        /* Only care about add/remove events for physical functions.
+         * For VFs we actually do the lookup based on the corresponding
+         * PF in device_to_iommu() anyway. */
         if (pdev->is_virtfn)
                 return NOTIFY_DONE;
         if (action != BUS_NOTIFY_ADD_DEVICE &&
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index a4407eabf0e6..2723090a0d54 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -892,7 +892,13 @@ static struct intel_iommu *device_to_iommu(struct device *dev, u8 *bus, u8 *devf
                 return NULL;
 
         if (dev_is_pci(dev)) {
+                struct pci_dev *pf_pdev;
+
                 pdev = to_pci_dev(dev);
+                /* VFs aren't listed in scope tables; we need to look up
+                 * the PF instead to find the IOMMU. */
+                pf_pdev = pci_physfn(pdev);
+                dev = &pf_pdev->dev;
                 segment = pci_domain_nr(pdev->bus);
         } else if (has_acpi_companion(dev))
                 dev = &ACPI_COMPANION(dev)->dev;
@@ -905,6 +911,13 @@ static struct intel_iommu *device_to_iommu(struct device *dev, u8 *bus, u8 *devf
                 for_each_active_dev_scope(drhd->devices,
                                           drhd->devices_cnt, i, tmp) {
                         if (tmp == dev) {
+                                /* For a VF use its original BDF# not that of the PF
+                                 * which we used for the IOMMU lookup. Strictly speaking
+                                 * we could do this for all PCI devices; we only need to
+                                 * get the BDF# from the scope table for ACPI matches. */
+                                if (pdev->is_virtfn)
+                                        goto got_pdev;
+
                                 *bus = drhd->devices[i].bus;
                                 *devfn = drhd->devices[i].devfn;
                                 goto out;