3 files changed, 47 insertions, 20 deletions

diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 235c0f4d3861..7250bef7f49e 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -75,6 +75,11 @@ DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info);
 enum xen_domain_type xen_domain_type = XEN_NATIVE;
 EXPORT_SYMBOL_GPL(xen_domain_type);
 
+unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START;
+EXPORT_SYMBOL(machine_to_phys_mapping);
+unsigned int   machine_to_phys_order;
+EXPORT_SYMBOL(machine_to_phys_order);
+
 struct start_info *xen_start_info;
 EXPORT_SYMBOL_GPL(xen_start_info);
 
@@ -1090,6 +1095,8 @@ static void __init xen_setup_stackprotector(void)
 /* First C function to be called on Xen boot */
 asmlinkage void __init xen_start_kernel(void)
 {
+        struct physdev_set_iopl set_iopl;
+        int rc;
         pgd_t *pgd;
 
         if (!xen_start_info)
@@ -1097,6 +1104,8 @@ asmlinkage void __init xen_start_kernel(void)
 
         xen_domain_type = XEN_PV_DOMAIN;
 
+        xen_setup_machphys_mapping();
+
         /* Install Xen paravirt ops */
         pv_info = xen_info;
         pv_init_ops = xen_init_ops;
@@ -1202,10 +1211,18 @@ asmlinkage void __init xen_start_kernel(void)
 #else
         pv_info.kernel_rpl = 0;
 #endif
-
         /* set the limit of our address space */
         xen_reserve_top();
 
+        /* We used to do this in xen_arch_setup, but that is too late on AMD
+         * were early_cpu_init (run before ->arch_setup()) calls early_amd_init
+         * which pokes 0xcf8 port.
+         */
+        set_iopl.iopl = 1;
+        rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
+        if (rc != 0)
+                xen_raw_printk("physdev_op failed %d\n", rc);
+
 #ifdef CONFIG_X86_32
         /* set up basic CPUID stuff */
         cpu_detect(&new_cpu_data);
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index c237b810b03f..790af908284e 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -2034,6 +2034,20 @@ static __init void xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn)
         set_page_prot(pmd, PAGE_KERNEL_RO);
 }
 
+void __init xen_setup_machphys_mapping(void)
+{
+        struct xen_machphys_mapping mapping;
+        unsigned long machine_to_phys_nr_ents;
+
+        if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) {
+                machine_to_phys_mapping = (unsigned long *)mapping.v_start;
+                machine_to_phys_nr_ents = mapping.max_mfn + 1;
+        } else {
+                machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES;
+        }
+        machine_to_phys_order = fls(machine_to_phys_nr_ents - 1);
+}
+
 #ifdef CONFIG_X86_64
 static void convert_pfn_mfn(void *v)
 {
@@ -2126,7 +2140,7 @@ __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd,
 {
         pmd_t *kernel_pmd;
 
-        level2_kernel_pgt = extend_brk(sizeof(pmd_t *) * PTRS_PER_PMD, PAGE_SIZE);
+        level2_kernel_pgt = extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE);
 
         max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->pt_base) +
                                   xen_start_info->nr_pt_frames * PAGE_SIZE +
@@ -2627,7 +2641,8 @@ int xen_remap_domain_mfn_range(struct vm_area_struct *vma,
 
         prot = __pgprot(pgprot_val(prot) | _PAGE_IOMAP);
 
-        vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
+        BUG_ON(!((vma->vm_flags & (VM_PFNMAP | VM_RESERVED | VM_IO)) ==
+                                (VM_PFNMAP | VM_RESERVED | VM_IO)));
 
         rmd.mfn = mfn;
         rmd.prot = prot;
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index b1dbdaa23ecc..38fdffaa71d3 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -118,16 +118,18 @@ static unsigned long __init xen_return_unused_memory(unsigned long max_pfn,
                                                      const struct e820map *e820)
 {
         phys_addr_t max_addr = PFN_PHYS(max_pfn);
-        phys_addr_t last_end = 0;
+        phys_addr_t last_end = ISA_END_ADDRESS;
         unsigned long released = 0;
         int i;
 
+        /* Free any unused memory above the low 1Mbyte. */
         for (i = 0; i < e820->nr_map && last_end < max_addr; i++) {
                 phys_addr_t end = e820->map[i].addr;
                 end = min(max_addr, end);
 
-                released += xen_release_chunk(last_end, end);
-                last_end = e820->map[i].addr + e820->map[i].size;
+                if (last_end < end)
+                        released += xen_release_chunk(last_end, end);
+                last_end = max(last_end, e820->map[i].addr + e820->map[i].size);
         }
 
         if (last_end < max_addr)
@@ -164,6 +166,7 @@ char * __init xen_memory_setup(void)
                 XENMEM_memory_map;
         rc = HYPERVISOR_memory_op(op, &memmap);
         if (rc == -ENOSYS) {
+                BUG_ON(xen_initial_domain());
                 memmap.nr_entries = 1;
                 map[0].addr = 0ULL;
                 map[0].size = mem_end;
@@ -201,12 +204,13 @@ char * __init xen_memory_setup(void)
         }
 
         /*
-         * Even though this is normal, usable memory under Xen, reserve
-         * ISA memory anyway because too many things think they can poke
+         * In domU, the ISA region is normal, usable memory, but we
+         * reserve ISA memory anyway because too many things poke
          * about in there.
          *
-         * In a dom0 kernel, this region is identity mapped with the
-         * hardware ISA area, so it really is out of bounds.
+         * In Dom0, the host E820 information can leave gaps in the
+         * ISA range, which would cause us to release those pages.  To
+         * avoid this, we unconditionally reserve them here.
          */
         e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS,
                         E820_RESERVED);
@@ -244,8 +248,7 @@ char * __init xen_memory_setup(void)
         else
                 extra_pages = 0;
 
-        if (!xen_initial_domain())
-                xen_add_extra_mem(extra_pages);
+        xen_add_extra_mem(extra_pages);
 
         return "Xen";
 }
@@ -333,9 +336,6 @@ void __cpuinit xen_enable_syscall(void)
 
 void __init xen_arch_setup(void)
 {
-        struct physdev_set_iopl set_iopl;
-        int rc;
-
         xen_panic_handler_init();
 
         HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_4gb_segments);
@@ -352,11 +352,6 @@ void __init xen_arch_setup(void)
         xen_enable_sysenter();
         xen_enable_syscall();
 
-        set_iopl.iopl = 1;
-        rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
-        if (rc != 0)
-                printk(KERN_INFO "physdev_op failed %d\n", rc);
-
 #ifdef CONFIG_ACPI
         if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
                 printk(KERN_INFO "ACPI in unprivileged domain disabled\n");