File 5a5e2d73-x86-Meltdown-band-aid-conditional.patch of Package xen.7317

# Commit e871e80c38547d9faefc6604532ba3e985e65873
# Date 2018-01-16 17:50:59 +0100
# Author Jan Beulich <jbeulich@suse.com>
# Committer Jan Beulich <jbeulich@suse.com>
x86: allow Meltdown band-aid to be disabled

First of all we don't need it on AMD systems. Additionally allow its use
to be controlled by command line option. For best backportability, this
intentionally doesn't use alternative instruction patching to achieve
the intended effect - while we likely want it, this will be later
follow-up.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>

--- a/docs/misc/xen-command-line.markdown
+++ b/docs/misc/xen-command-line.markdown
@@ -1147,6 +1147,18 @@ In the case that x2apic is in use, this
 clustered mode.  The default, given no hint from the **FADT**, is cluster
 mode.
 
+### xpti
+> `= <boolean>`
+
+> Default: `false` on AMD hardware
+> Default: `true` everywhere else
+
+Override default selection of whether to isolate 64-bit PV guest page
+tables.
+
+** WARNING: Not yet a complete isolation implementation, but better than
+nothing. **
+
 ### xsave
 > `= <boolean>`
 
--- a/xen/arch/x86/domain.c
+++ b/xen/arch/x86/domain.c
@@ -1374,13 +1374,16 @@ static void paravirt_ctxt_switch_from(st
 
 static void paravirt_ctxt_switch_to(struct vcpu *v)
 {
+    root_pgentry_t *root_pgt = this_cpu(root_pgt);
     unsigned long cr4;
 
     set_int80_direct_trap(v);
     switch_kernel_stack(v);
 
-    this_cpu(root_pgt)[root_table_offset(PERDOMAIN_VIRT_START)] =
-        l4e_from_page(v->domain->arch.perdomain_l3_pg, __PAGE_HYPERVISOR);
+    if ( root_pgt )
+        root_pgt[root_table_offset(PERDOMAIN_VIRT_START)] =
+            l4e_from_page(v->domain->arch.perdomain_l3_pg,
+                          __PAGE_HYPERVISOR);
 
     cr4 = pv_guest_cr4_to_real_cr4(v);
     if ( unlikely(cr4 != read_cr4()) )
--- a/xen/arch/x86/mm.c
+++ b/xen/arch/x86/mm.c
@@ -3822,7 +3822,7 @@ long do_mmu_update(
                     rc = mod_l4_entry(va, l4e_from_intpte(req.val), mfn,
                                       cmd == MMU_PT_UPDATE_PRESERVE_AD, v);
                     if ( !rc )
-                        sync_guest = 1;
+                        sync_guest = !!this_cpu(root_pgt);
                 break;
                 case PGT_writable_page:
                     perfc_incr(writable_mmu_updates);
--- a/xen/arch/x86/smpboot.c
+++ b/xen/arch/x86/smpboot.c
@@ -349,7 +349,7 @@ void start_secondary(void *unused)
     spin_debug_disable();
 
     get_cpu_info()->xen_cr3 = 0;
-    get_cpu_info()->pv_cr3 = __pa(this_cpu(root_pgt));
+    get_cpu_info()->pv_cr3 = this_cpu(root_pgt) ? __pa(this_cpu(root_pgt)) : 0;
 
     percpu_traps_init();
 
@@ -736,14 +736,20 @@ static int clone_mapping(const void *ptr
     return 0;
 }
 
+static __read_mostly int8_t opt_xpti = -1;
+boolean_param("xpti", opt_xpti);
 DEFINE_PER_CPU(root_pgentry_t *, root_pgt);
 
 static int setup_cpu_root_pgt(unsigned int cpu)
 {
-    root_pgentry_t *rpt = alloc_xen_pagetable();
+    root_pgentry_t *rpt;
     unsigned int off;
     int rc;
 
+    if ( !opt_xpti )
+        return 0;
+
+    rpt = alloc_xen_pagetable();
     if ( !rpt )
         return -ENOMEM;
 
@@ -935,10 +941,14 @@ void __init smp_prepare_cpus(unsigned in
 
     stack_base[0] = stack_start;
 
+    if ( opt_xpti < 0 )
+        opt_xpti = boot_cpu_data.x86_vendor != X86_VENDOR_AMD;
+
     rc = setup_cpu_root_pgt(0);
     if ( rc )
         panic("Error %d setting up PV root page table\n", rc);
-    get_cpu_info()->pv_cr3 = __pa(per_cpu(root_pgt, 0));
+    if ( per_cpu(root_pgt, 0) )
+        get_cpu_info()->pv_cr3 = __pa(per_cpu(root_pgt, 0));
 
     if ( !zalloc_cpumask_var(&per_cpu(cpu_sibling_mask, 0)) ||
          !zalloc_cpumask_var(&per_cpu(cpu_core_mask, 0)) )
@@ -999,6 +1009,7 @@ void __init smp_prepare_boot_cpu(void)
     cpumask_set_cpu(smp_processor_id(), &cpu_present_map);
 
     get_cpu_info()->xen_cr3 = 0;
+    get_cpu_info()->pv_cr3 = 0;
 }
 
 static void
--- a/xen/arch/x86/x86_64/entry.S
+++ b/xen/arch/x86/x86_64/entry.S
@@ -46,6 +46,7 @@ restore_all_guest:
         movabs $DIRECTMAP_VIRT_START, %rcx
         mov   %rdi, %rax
         and   %rsi, %rdi
+        jz    .Lrag_keep_cr3
         and   %r9, %rsi
         add   %rcx, %rdi
         add   %rcx, %rsi
@@ -62,6 +63,7 @@ restore_all_guest:
         rep movsq
         mov   %r9, STACK_CPUINFO_FIELD(xen_cr3)(%rdx)
         write_cr3 rax, rdi, rsi
+.Lrag_keep_cr3:
 
         RESTORE_ALL
         testw $TRAP_syscall,4(%rsp)