x86/mm/p2m: stop checking for IOMMU shared page tables in mmio_order()

Now that the iommu_map() and iommu_unmap() operations take an order
parameter and elide flushing there's no strong reason why modifying MMIO
ranges in the p2m should be restricted to a 4k granularity simply because
the IOMMU is enabled but shared page tables are not in operation.

Signed-off-by: Paul Durrant <[email protected]>
Reviewed-by: Jan Beulich <[email protected]>

Author: Paul Durrant

xen/arch/x86/mm/p2m.c

@@ -2210,13 +2210,12 @@ static unsigned int mmio_order(const struct domain *d,
                                unsigned long start_fn, unsigned long nr)
 {
     /*
-     * Note that the !iommu_use_hap_pt() here has three effects:
-     * - cover iommu_{,un}map_page() not having an "order" input yet,
+     * Note that the !hap_enabled() here has two effects:
      * - exclude shadow mode (which doesn't support large MMIO mappings),
      * - exclude PV guests, should execution reach this code for such.
      * So be careful when altering this.
      */
-    if ( !iommu_use_hap_pt(d) ||
+    if ( !hap_enabled(d) ||
          (start_fn & ((1UL << PAGE_ORDER_2M) - 1)) || !(nr >> PAGE_ORDER_2M) )
         return PAGE_ORDER_4K;