[xen master] amd/iommu: remove hidden AMD inclusive mappings

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[xen master] amd/iommu: remove hidden AMD inclusive mappings

patchbot
commit d677097dc12c15785db8d988253837d0ee7347f4
Author:     Roger Pau Monné <[hidden email]>
AuthorDate: Tue Oct 2 17:02:33 2018 +0200
Commit:     Jan Beulich <[hidden email]>
CommitDate: Tue Oct 2 17:02:33 2018 +0200

    amd/iommu: remove hidden AMD inclusive mappings
   
    And just rely on arch_iommu_hwdom_init to setup the correct inclusive
    mappings as it's done for Intel.
   
    AMD has code in amd_iommu_hwdom_init to setup inclusive mappings up to
    max_pdx, remove this since it's now a duplication of
    arch_iommu_hwdom_init. Note that AMD mapped every page with a valid
    mfn up to max_pdx, arch_iommu_hwdom_init will only do so for memory
    below 4GB, so this is a functional change for AMD.
   
    Move the default setting of iommu_hwdom_{inclusive/reserved} to
    arch_iommu_hwdom_init since the defaults are now the same for both
    Intel and AMD.
   
    Reported-by: Paul Durrant <[hidden email]>
    Signed-off-by: Roger Pau Monné <[hidden email]>
    Reviewed-by: Jan Beulich <[hidden email]>
    Reviewed-by: Kevin Tian <[hidden email]>
    Acked-by: Suravee Suthikulpanit <[hidden email]>
---
 xen/drivers/passthrough/amd/pci_amd_iommu.c | 39 -----------------------------
 xen/drivers/passthrough/vtd/iommu.c         |  7 ------
 xen/drivers/passthrough/x86/iommu.c         |  8 +++++-
 3 files changed, 7 insertions(+), 47 deletions(-)

diff --git a/xen/drivers/passthrough/amd/pci_amd_iommu.c b/xen/drivers/passthrough/amd/pci_amd_iommu.c
index 4a633ca940..821fe03df5 100644
--- a/xen/drivers/passthrough/amd/pci_amd_iommu.c
+++ b/xen/drivers/passthrough/amd/pci_amd_iommu.c
@@ -250,50 +250,11 @@ static int amd_iommu_add_device(u8 devfn, struct pci_dev *pdev);
 
 static void __hwdom_init amd_iommu_hwdom_init(struct domain *d)
 {
-    unsigned long i;
     const struct amd_iommu *iommu;
 
-    /* Inclusive IOMMU mappings are disabled by default on AMD hardware. */
-    if ( iommu_hwdom_inclusive == -1 )
-        iommu_hwdom_inclusive = 0;
-    /* Reserved IOMMU mappings are disabled by default on AMD hardware. */
-    if ( iommu_hwdom_reserved == -1 )
-        iommu_hwdom_reserved = 0;
-
     if ( allocate_domain_resources(dom_iommu(d)) )
         BUG();
 
-    if ( !iommu_hwdom_passthrough && !need_iommu(d) )
-    {
-        int rc = 0;
-
-        /* Set up 1:1 page table for dom0 */
-        for ( i = 0; i < max_pdx; i++ )
-        {
-            unsigned long pfn = pdx_to_pfn(i);
-
-            /*
-             * XXX Should we really map all non-RAM (above 4G)? Minimally
-             * a pfn_valid() check would seem desirable here.
-             */
-            if ( mfn_valid(_mfn(pfn)) )
-            {
-                int ret = amd_iommu_map_page(d, pfn, pfn,
-                                             IOMMUF_readable|IOMMUF_writable);
-
-                if ( !rc )
-                    rc = ret;
-            }
-
-            if ( !(i & 0xfffff) )
-                process_pending_softirqs();
-        }
-
-        if ( rc )
-            AMD_IOMMU_DEBUG("d%d: IOMMU mapping failed: %d\n",
-                            d->domain_id, rc);
-    }
-
     for_each_amd_iommu ( iommu )
         if ( iomem_deny_access(d, PFN_DOWN(iommu->mmio_base_phys),
                                PFN_DOWN(iommu->mmio_base_phys +
diff --git a/xen/drivers/passthrough/vtd/iommu.c b/xen/drivers/passthrough/vtd/iommu.c
index bb422ec58c..cf8a80d7a1 100644
--- a/xen/drivers/passthrough/vtd/iommu.c
+++ b/xen/drivers/passthrough/vtd/iommu.c
@@ -1304,13 +1304,6 @@ static void __hwdom_init intel_iommu_hwdom_init(struct domain *d)
 {
     struct acpi_drhd_unit *drhd;
 
-    /* Inclusive mappings are enabled by default on Intel hardware for PV. */
-    if ( iommu_hwdom_inclusive == -1 )
-        iommu_hwdom_inclusive = is_pv_domain(d);
-    /* Reserved IOMMU mappings are enabled by default on Intel hardware. */
-    if ( iommu_hwdom_reserved == -1 )
-        iommu_hwdom_reserved = 1;
-
     setup_hwdom_pci_devices(d, setup_hwdom_device);
     setup_hwdom_rmrr(d);
     /* Make sure workarounds are applied before enabling the IOMMU(s). */
diff --git a/xen/drivers/passthrough/x86/iommu.c b/xen/drivers/passthrough/x86/iommu.c
index 69e45b8e00..2de8822c59 100644
--- a/xen/drivers/passthrough/x86/iommu.c
+++ b/xen/drivers/passthrough/x86/iommu.c
@@ -210,7 +210,13 @@ void __hwdom_init arch_iommu_hwdom_init(struct domain *d)
 
     BUG_ON(!is_hardware_domain(d));
 
-    ASSERT(iommu_hwdom_inclusive != -1 && iommu_hwdom_reserved != -1);
+    /* Inclusive mappings are enabled by default for PV. */
+    if ( iommu_hwdom_inclusive == -1 )
+        iommu_hwdom_inclusive = is_pv_domain(d);
+    /* Reserved IOMMU mappings are enabled by default. */
+    if ( iommu_hwdom_reserved == -1 )
+        iommu_hwdom_reserved = 1;
+
     if ( iommu_hwdom_inclusive && !is_pv_domain(d) )
     {
         printk(XENLOG_WARNING
--
generated by git-patchbot for /home/xen/git/xen.git#master


_______________________________________________
Xen-changelog mailing list
[hidden email]
https://lists.xenproject.org/xen-changelog