[Xen-devel] [PATCH v5 6/6] amd/iommu: skip bridge devices when updating IOMMU page tables

[Roger Pau Monne <roger.pau@xxxxxxxxxx>]

Bridges are not behind an IOMMU, and are already special cased and
skipped in amd_iommu_add_device. Apply the same special casing when
updating page tables.

This is required or else update_paging_mode will fail and return an
error to the caller (amd_iommu_{un}map_page) which will destroy the
domain.

Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
---
Cc: Suravee Suthikulpanit <suravee.suthikulpanit@xxxxxxx>
Cc: Brian Woods <brian.woods@xxxxxxx>
---
Changes since v4:
 - Invert condition order so they match the order in
   amd_iommu_add_device.
 - Expand commit message to spell out why this is required.
---
 xen/drivers/passthrough/amd/iommu_map.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/xen/drivers/passthrough/amd/iommu_map.c 
b/xen/drivers/passthrough/amd/iommu_map.c
index c1daba8422..7752e7542f 100644
--- a/xen/drivers/passthrough/amd/iommu_map.c
+++ b/xen/drivers/passthrough/amd/iommu_map.c
@@ -612,6 +612,10 @@ static int update_paging_mode(struct domain *d, unsigned 
long dfn)
         /* Update device table entries using new root table and paging mode */
         for_each_pdev( d, pdev )
         {
+            if ( pdev->type == DEV_TYPE_PCI_HOST_BRIDGE &&
+                 is_hardware_domain(d) )
+                continue;
+
             bdf = PCI_BDF2(pdev->bus, pdev->devfn);
             iommu = find_iommu_for_device(pdev->seg, bdf);
             if ( !iommu )
-- 
2.19.1


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel