Merge branch 'linus' into stackprotector

Conflicts:
	arch/x86/kernel/Makefile
	include/asm-x86/pda.h

20 files changed, 3066 insertions, 754 deletions

diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig
index 6c388e593bc..87b9ab16642 100644
--- a/arch/x86/xen/Kconfig
+++ b/arch/x86/xen/Kconfig
@@ -6,9 +6,33 @@ config XEN
 	bool "Xen guest support"
 	select PARAVIRT
 	select PARAVIRT_CLOCK
-	depends on X86_32
-	depends on X86_CMPXCHG && X86_TSC && X86_PAE && !(X86_VISWS || X86_VOYAGER)
+	depends on X86_64 || (X86_32 && X86_PAE && !(X86_VISWS || X86_VOYAGER))
+	depends on X86_CMPXCHG && X86_TSC
 	help
 	  This is the Linux Xen port.  Enabling this will allow the
 	  kernel to boot in a paravirtualized environment under the
 	  Xen hypervisor.
+
+config XEN_MAX_DOMAIN_MEMORY
+       int "Maximum allowed size of a domain in gigabytes"
+       default 8 if X86_32
+       default 32 if X86_64
+       depends on XEN
+       help
+         The pseudo-physical to machine address array is sized
+         according to the maximum possible memory size of a Xen
+         domain.  This array uses 1 page per gigabyte, so there's no
+         need to be too stingy here.
+
+config XEN_SAVE_RESTORE
+       bool
+       depends on XEN && PM
+       default y
+
+config XEN_DEBUG_FS
+	bool "Enable Xen debug and tuning parameters in debugfs"
+	depends on XEN && DEBUG_FS
+	default n
+	help
+	  Enable statistics output and various tuning options in debugfs.
+	  Enabling this option may incur a significant performance overhead.
diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile
index 3d8df981d5f..313947940a1 100644
--- a/arch/x86/xen/Makefile
+++ b/arch/x86/xen/Makefile
@@ -1,4 +1,12 @@
-obj-y		:= enlighten.o setup.o multicalls.o mmu.o \
-			time.o manage.o xen-asm.o grant-table.o
+ifdef CONFIG_FTRACE
+# Do not profile debug and lowlevel utilities
+CFLAGS_REMOVE_spinlock.o = -pg
+CFLAGS_REMOVE_time.o = -pg
+CFLAGS_REMOVE_irq.o = -pg
+endif
 
-obj-$(CONFIG_SMP)	+= smp.o
+obj-y		:= enlighten.o setup.o multicalls.o mmu.o irq.o \
+			time.o xen-asm_$(BITS).o grant-table.o suspend.o
+
+obj-$(CONFIG_SMP)		+= smp.o spinlock.o
+obj-$(CONFIG_XEN_DEBUG_FS)	+= debugfs.o
\ No newline at end of file
diff --git a/arch/x86/xen/debugfs.c b/arch/x86/xen/debugfs.c
new file mode 100644
index 00000000000..b53225d2cac
--- /dev/null
+++ b/arch/x86/xen/debugfs.c
@@ -0,0 +1,123 @@
+#include <linux/init.h>
+#include <linux/debugfs.h>
+#include <linux/module.h>
+
+#include "debugfs.h"
+
+static struct dentry *d_xen_debug;
+
+struct dentry * __init xen_init_debugfs(void)
+{
+	if (!d_xen_debug) {
+		d_xen_debug = debugfs_create_dir("xen", NULL);
+
+		if (!d_xen_debug)
+			pr_warning("Could not create 'xen' debugfs directory\n");
+	}
+
+	return d_xen_debug;
+}
+
+struct array_data
+{
+	void *array;
+	unsigned elements;
+};
+
+static int u32_array_open(struct inode *inode, struct file *file)
+{
+	file->private_data = NULL;
+	return nonseekable_open(inode, file);
+}
+
+static size_t format_array(char *buf, size_t bufsize, const char *fmt,
+			   u32 *array, unsigned array_size)
+{
+	size_t ret = 0;
+	unsigned i;
+
+	for(i = 0; i < array_size; i++) {
+		size_t len;
+
+		len = snprintf(buf, bufsize, fmt, array[i]);
+		len++;	/* ' ' or '\n' */
+		ret += len;
+
+		if (buf) {
+			buf += len;
+			bufsize -= len;
+			buf[-1] = (i == array_size-1) ? '\n' : ' ';
+		}
+	}
+
+	ret++;		/* \0 */
+	if (buf)
+		*buf = '\0';
+
+	return ret;
+}
+
+static char *format_array_alloc(const char *fmt, u32 *array, unsigned array_size)
+{
+	size_t len = format_array(NULL, 0, fmt, array, array_size);
+	char *ret;
+
+	ret = kmalloc(len, GFP_KERNEL);
+	if (ret == NULL)
+		return NULL;
+
+	format_array(ret, len, fmt, array, array_size);
+	return ret;
+}
+
+static ssize_t u32_array_read(struct file *file, char __user *buf, size_t len,
+			      loff_t *ppos)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct array_data *data = inode->i_private;
+	size_t size;
+
+	if (*ppos == 0) {
+		if (file->private_data) {
+			kfree(file->private_data);
+			file->private_data = NULL;
+		}
+
+		file->private_data = format_array_alloc("%u", data->array, data->elements);
+	}
+
+	size = 0;
+	if (file->private_data)
+		size = strlen(file->private_data);
+
+	return simple_read_from_buffer(buf, len, ppos, file->private_data, size);
+}
+
+static int xen_array_release(struct inode *inode, struct file *file)
+{
+	kfree(file->private_data);
+
+	return 0;
+}
+
+static struct file_operations u32_array_fops = {
+	.owner	= THIS_MODULE,
+	.open	= u32_array_open,
+	.release= xen_array_release,
+	.read	= u32_array_read,
+};
+
+struct dentry *xen_debugfs_create_u32_array(const char *name, mode_t mode,
+					    struct dentry *parent,
+					    u32 *array, unsigned elements)
+{
+	struct array_data *data = kmalloc(sizeof(*data), GFP_KERNEL);
+
+	if (data == NULL)
+		return NULL;
+
+	data->array = array;
+	data->elements = elements;
+
+	return debugfs_create_file(name, mode, parent, data, &u32_array_fops);
+}
diff --git a/arch/x86/xen/debugfs.h b/arch/x86/xen/debugfs.h
new file mode 100644
index 00000000000..e2813208483
--- /dev/null
+++ b/arch/x86/xen/debugfs.h
@@ -0,0 +1,10 @@
+#ifndef _XEN_DEBUGFS_H
+#define _XEN_DEBUGFS_H
+
+struct dentry * __init xen_init_debugfs(void);
+
+struct dentry *xen_debugfs_create_u32_array(const char *name, mode_t mode,
+					    struct dentry *parent,
+					    u32 *array, unsigned elements);
+
+#endif /* _XEN_DEBUGFS_H */
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index f09c1c69c37..0013a729b41 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -30,16 +30,18 @@
 #include <xen/interface/xen.h>
 #include <xen/interface/physdev.h>
 #include <xen/interface/vcpu.h>
-#include <xen/interface/sched.h>
 #include <xen/features.h>
 #include <xen/page.h>
+#include <xen/hvc-console.h>
 
 #include <asm/paravirt.h>
+#include <asm/apic.h>
 #include <asm/page.h>
 #include <asm/xen/hypercall.h>
 #include <asm/xen/hypervisor.h>
 #include <asm/fixmap.h>
 #include <asm/processor.h>
+#include <asm/msr-index.h>
 #include <asm/setup.h>
 #include <asm/desc.h>
 #include <asm/pgtable.h>
@@ -55,6 +57,21 @@ EXPORT_SYMBOL_GPL(hypercall_page);
 DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu);
 DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info);
 
+enum xen_domain_type xen_domain_type = XEN_NATIVE;
+EXPORT_SYMBOL_GPL(xen_domain_type);
+
+/*
+ * Identity map, in addition to plain kernel map.  This needs to be
+ * large enough to allocate page table pages to allocate the rest.
+ * Each page can map 2MB.
+ */
+static pte_t level1_ident_pgt[PTRS_PER_PTE * 4] __page_aligned_bss;
+
+#ifdef CONFIG_X86_64
+/* l3 pud for userspace vsyscall mapping */
+static pud_t level3_user_vsyscall[PTRS_PER_PUD] __page_aligned_bss;
+#endif /* CONFIG_X86_64 */
+
 /*
  * Note about cr3 (pagetable base) values:
  *
@@ -75,13 +92,13 @@ DEFINE_PER_CPU(unsigned long, xen_current_cr3);	 /* actual vcpu cr3 */
 struct start_info *xen_start_info;
 EXPORT_SYMBOL_GPL(xen_start_info);
 
-static /* __initdata */ struct shared_info dummy_shared_info;
+struct shared_info xen_dummy_shared_info;
 
 /*
  * Point at some empty memory to start with. We map the real shared_info
  * page as soon as fixmap is up and running.
  */
-struct shared_info *HYPERVISOR_shared_info = (void *)&dummy_shared_info;
+struct shared_info *HYPERVISOR_shared_info = (void *)&xen_dummy_shared_info;
 
 /*
  * Flag to determine whether vcpu info placement is available on all
@@ -96,15 +113,22 @@ struct shared_info *HYPERVISOR_shared_info = (void *)&dummy_shared_info;
  *
  * 0: not available, 1: available
  */
-static int have_vcpu_info_placement = 1;
+static int have_vcpu_info_placement =
+#ifdef CONFIG_X86_32
+	1
+#else
+	0
+#endif
+	;
 
-static void __init xen_vcpu_setup(int cpu)
+
+static void xen_vcpu_setup(int cpu)
 {
 	struct vcpu_register_vcpu_info info;
 	int err;
 	struct vcpu_info *vcpup;
 
-	BUG_ON(HYPERVISOR_shared_info == &dummy_shared_info);
+	BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
 	per_cpu(xen_vcpu, cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu];
 
 	if (!have_vcpu_info_placement)
@@ -136,11 +160,45 @@ static void __init xen_vcpu_setup(int cpu)
 	}
 }
 
+/*
+ * On restore, set the vcpu placement up again.
+ * If it fails, then we're in a bad state, since
+ * we can't back out from using it...
+ */
+void xen_vcpu_restore(void)
+{
+	if (have_vcpu_info_placement) {
+		int cpu;
+
+		for_each_online_cpu(cpu) {
+			bool other_cpu = (cpu != smp_processor_id());
+
+			if (other_cpu &&
+			    HYPERVISOR_vcpu_op(VCPUOP_down, cpu, NULL))
+				BUG();
+
+			xen_vcpu_setup(cpu);
+
+			if (other_cpu &&
+			    HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL))
+				BUG();
+		}
+
+		BUG_ON(!have_vcpu_info_placement);
+	}
+}
+
 static void __init xen_banner(void)
 {
+	unsigned version = HYPERVISOR_xen_version(XENVER_version, NULL);
+	struct xen_extraversion extra;
+	HYPERVISOR_xen_version(XENVER_extraversion, &extra);
+
 	printk(KERN_INFO "Booting paravirtualized kernel on %s\n",
 	       pv_info.name);
-	printk(KERN_INFO "Hypervisor signature: %s\n", xen_start_info->magic);
+	printk(KERN_INFO "Xen version: %d.%d%s%s\n",
+	       version >> 16, version & 0xffff, extra.extraversion,
+	       xen_feature(XENFEAT_mmu_pt_update_preserve_ad) ? " (preserve-AD)" : "");
 }
 
 static void xen_cpuid(unsigned int *ax, unsigned int *bx,
@@ -178,103 +236,68 @@ static unsigned long xen_get_debugreg(int reg)
 	return HYPERVISOR_get_debugreg(reg);
 }
 
-static unsigned long xen_save_fl(void)
-{
-	struct vcpu_info *vcpu;
-	unsigned long flags;
-
-	vcpu = x86_read_percpu(xen_vcpu);
-
-	/* flag has opposite sense of mask */
-	flags = !vcpu->evtchn_upcall_mask;
-
-	/* convert to IF type flag
-	   -0 -> 0x00000000
-	   -1 -> 0xffffffff
-	*/
-	return (-flags) & X86_EFLAGS_IF;
-}
-
-static void xen_restore_fl(unsigned long flags)
+static void xen_leave_lazy(void)
 {
-	struct vcpu_info *vcpu;
-
-	/* convert from IF type flag */
-	flags = !(flags & X86_EFLAGS_IF);
-
-	/* There's a one instruction preempt window here.  We need to
-	   make sure we're don't switch CPUs between getting the vcpu
-	   pointer and updating the mask. */
-	preempt_disable();
-	vcpu = x86_read_percpu(xen_vcpu);
-	vcpu->evtchn_upcall_mask = flags;
-	preempt_enable_no_resched();
-
-	/* Doesn't matter if we get preempted here, because any
-	   pending event will get dealt with anyway. */
-
-	if (flags == 0) {
-		preempt_check_resched();
-		barrier(); /* unmask then check (avoid races) */
-		if (unlikely(vcpu->evtchn_upcall_pending))
-			force_evtchn_callback();
-	}
+	paravirt_leave_lazy(paravirt_get_lazy_mode());
+	xen_mc_flush();
 }
 
-static void xen_irq_disable(void)
+static unsigned long xen_store_tr(void)
 {
-	/* There's a one instruction preempt window here.  We need to
-	   make sure we're don't switch CPUs between getting the vcpu
-	   pointer and updating the mask. */
-	preempt_disable();
-	x86_read_percpu(xen_vcpu)->evtchn_upcall_mask = 1;
-	preempt_enable_no_resched();
+	return 0;
 }
 
-static void xen_irq_enable(void)
+/*
+ * Set the page permissions for a particular virtual address.  If the
+ * address is a vmalloc mapping (or other non-linear mapping), then
+ * find the linear mapping of the page and also set its protections to
+ * match.
+ */
+static void set_aliased_prot(void *v, pgprot_t prot)
 {
-	struct vcpu_info *vcpu;
+	int level;
+	pte_t *ptep;
+	pte_t pte;
+	unsigned long pfn;
+	struct page *page;
 
-	/* There's a one instruction preempt window here.  We need to
-	   make sure we're don't switch CPUs between getting the vcpu
-	   pointer and updating the mask. */
-	preempt_disable();
-	vcpu = x86_read_percpu(xen_vcpu);
-	vcpu->evtchn_upcall_mask = 0;
-	preempt_enable_no_resched();
+	ptep = lookup_address((unsigned long)v, &level);
+	BUG_ON(ptep == NULL);
 
-	/* Doesn't matter if we get preempted here, because any
-	   pending event will get dealt with anyway. */
+	pfn = pte_pfn(*ptep);
+	page = pfn_to_page(pfn);
 
-	barrier(); /* unmask then check (avoid races) */
-	if (unlikely(vcpu->evtchn_upcall_pending))
-		force_evtchn_callback();
-}
+	pte = pfn_pte(pfn, prot);
 
-static void xen_safe_halt(void)
-{
-	/* Blocking includes an implicit local_irq_enable(). */
-	if (HYPERVISOR_sched_op(SCHEDOP_block, 0) != 0)
+	if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0))
 		BUG();
-}
 
-static void xen_halt(void)
-{
-	if (irqs_disabled())
-		HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
-	else
-		xen_safe_halt();
+	if (!PageHighMem(page)) {
+		void *av = __va(PFN_PHYS(pfn));
+
+		if (av != v)
+			if (HYPERVISOR_update_va_mapping((unsigned long)av, pte, 0))
+				BUG();
+	} else
+		kmap_flush_unused();
 }
 
-static void xen_leave_lazy(void)
+static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries)
 {
-	paravirt_leave_lazy(paravirt_get_lazy_mode());
-	xen_mc_flush();
+	const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE;
+	int i;
+
+	for(i = 0; i < entries; i += entries_per_page)
+		set_aliased_prot(ldt + i, PAGE_KERNEL_RO);
 }
 
-static unsigned long xen_store_tr(void)
+static void xen_free_ldt(struct desc_struct *ldt, unsigned entries)
 {
-	return 0;
+	const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE;
+	int i;
+
+	for(i = 0; i < entries; i += entries_per_page)
+		set_aliased_prot(ldt + i, PAGE_KERNEL);
 }
 
 static void xen_set_ldt(const void *addr, unsigned entries)
@@ -332,14 +355,6 @@ static void load_TLS_descriptor(struct thread_struct *t,
 
 static void xen_load_tls(struct thread_struct *t, unsigned int cpu)
 {
-	xen_mc_batch();
-
-	load_TLS_descriptor(t, cpu, 0);
-	load_TLS_descriptor(t, cpu, 1);
-	load_TLS_descriptor(t, cpu, 2);
-
-	xen_mc_issue(PARAVIRT_LAZY_CPU);
-
 	/*
 	 * XXX sleazy hack: If we're being called in a lazy-cpu zone,
 	 * it means we're in a context switch, and %gs has just been
@@ -348,16 +363,44 @@ static void xen_load_tls(struct thread_struct *t, unsigned int cpu)
 	 * Either way, it has been saved, and the new value will get
 	 * loaded properly.  This will go away as soon as Xen has been
 	 * modified to not save/restore %gs for normal hypercalls.
+	 *
+	 * On x86_64, this hack is not used for %gs, because gs points
+	 * to KERNEL_GS_BASE (and uses it for PDA references), so we
+	 * must not zero %gs on x86_64
+	 *
+	 * For x86_64, we need to zero %fs, otherwise we may get an
+	 * exception between the new %fs descriptor being loaded and
+	 * %fs being effectively cleared at __switch_to().
 	 */
-	if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU)
+	if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) {
+#ifdef CONFIG_X86_32
 		loadsegment(gs, 0);
+#else
+		loadsegment(fs, 0);
+#endif
+	}
+
+	xen_mc_batch();
+
+	load_TLS_descriptor(t, cpu, 0);
+	load_TLS_descriptor(t, cpu, 1);
+	load_TLS_descriptor(t, cpu, 2);
+
+	xen_mc_issue(PARAVIRT_LAZY_CPU);
+}
+
+#ifdef CONFIG_X86_64
+static void xen_load_gs_index(unsigned int idx)
+{
+	if (HYPERVISOR_set_segment_base(SEGBASE_GS_USER_SEL, idx))
+		BUG();
 }
+#endif
 
 static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum,
 				const void *ptr)
 {
-	unsigned long lp = (unsigned long)&dt[entrynum];
-	xmaddr_t mach_lp = virt_to_machine(lp);
+	xmaddr_t mach_lp = arbitrary_virt_to_machine(&dt[entrynum]);
 	u64 entry = *(u64 *)ptr;
 
 	preempt_disable();
@@ -369,23 +412,18 @@ static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum,
 	preempt_enable();
 }
 
-static int cvt_gate_to_trap(int vector, u32 low, u32 high,
+static int cvt_gate_to_trap(int vector, const gate_desc *val,
 			    struct trap_info *info)
 {
-	u8 type, dpl;
-
-	type = (high >> 8) & 0x1f;
-	dpl = (high >> 13) & 3;
-
-	if (type != 0xf && type != 0xe)
+	if (val->type != 0xf && val->type != 0xe)
 		return 0;
 
 	info->vector = vector;
-	info->address = (high & 0xffff0000) | (low & 0x0000ffff);
-	info->cs = low >> 16;
-	info->flags = dpl;
+	info->address = gate_offset(*val);
+	info->cs = gate_segment(*val);
+	info->flags = val->dpl;
 	/* interrupt gates clear IF */
-	if (type == 0xe)
+	if (val->type == 0xe)
 		info->flags |= 4;
 
 	return 1;
@@ -412,11 +450,10 @@ static void xen_write_idt_entry(gate_desc *dt, int entrynum, const gate_desc *g)
 
 	if (p >= start && (p + 8) <= end) {
 		struct trap_info info[2];
-		u32 *desc = (u32 *)g;
 
 		info[1].address = 0;
 
-		if (cvt_gate_to_trap(entrynum, desc[0], desc[1], &info[0]))
+		if (cvt_gate_to_trap(entrynum, g, &info[0]))
 			if (HYPERVISOR_set_trap_table(info))
 				BUG();
 	}
@@ -429,13 +466,13 @@ static void xen_convert_trap_info(const struct desc_ptr *desc,
 {
 	unsigned in, out, count;
 
-	count = (desc->size+1) / 8;
+	count = (desc->size+1) / sizeof(gate_desc);
 	BUG_ON(count > 256);
 
 	for (in = out = 0; in < count; in++) {
-		const u32 *entry = (u32 *)(desc->address + in * 8);
+		gate_desc *entry = (gate_desc*)(desc->address) + in;
 
-		if (cvt_gate_to_trap(in, entry[0], entry[1], &traps[out]))
+		if (cvt_gate_to_trap(in, entry, &traps[out]))
 			out++;
 	}
 	traps[out].address = 0;
@@ -496,7 +533,7 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
 }
 
 static void xen_load_sp0(struct tss_struct *tss,
-			  struct thread_struct *thread)
+			 struct thread_struct *thread)
 {
 	struct multicall_space mcs = xen_mc_entry(0);
 	MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0);
@@ -517,16 +554,47 @@ static void xen_io_delay(void)
 }
 
 #ifdef CONFIG_X86_LOCAL_APIC
-static u32 xen_apic_read(unsigned long reg)
+static u32 xen_apic_read(u32 reg)
 {
 	return 0;
 }
 
-static void xen_apic_write(unsigned long reg, u32 val)
+static void xen_apic_write(u32 reg, u32 val)
 {
 	/* Warn to see if there's any stray references */
 	WARN_ON(1);
 }
+
+static u64 xen_apic_icr_read(void)
+{
+	return 0;
+}
+
+static void xen_apic_icr_write(u32 low, u32 id)
+{
+	/* Warn to see if there's any stray references */
+	WARN_ON(1);
+}
+
+static void xen_apic_wait_icr_idle(void)
+{
+        return;
+}
+
+static u32 xen_safe_apic_wait_icr_idle(void)
+{
+        return 0;
+}
+
+static struct apic_ops xen_basic_apic_ops = {
+	.read = xen_apic_read,
+	.write = xen_apic_write,
+	.icr_read = xen_apic_icr_read,
+	.icr_write = xen_apic_icr_write,
+	.wait_icr_idle = xen_apic_wait_icr_idle,
+	.safe_wait_icr_idle = xen_safe_apic_wait_icr_idle,
+};
+
 #endif
 
 static void xen_flush_tlb(void)
@@ -607,6 +675,30 @@ static void xen_flush_tlb_others(const cpumask_t *cpus, struct mm_struct *mm,
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
 }
 
+static void xen_clts(void)
+{
+	struct multicall_space mcs;
+
+	mcs = xen_mc_entry(0);
+
+	MULTI_fpu_taskswitch(mcs.mc, 0);
+
+	xen_mc_issue(PARAVIRT_LAZY_CPU);
+}
+
+static void xen_write_cr0(unsigned long cr0)
+{
+	struct multicall_space mcs;
+
+	/* Only pay attention to cr0.TS; everything else is
+	   ignored. */
+	mcs = xen_mc_entry(0);
+
+	MULTI_fpu_taskswitch(mcs.mc, (cr0 & X86_CR0_TS) != 0);
+
+	xen_mc_issue(PARAVIRT_LAZY_CPU);
+}
+
 static void xen_write_cr2(unsigned long cr2)
 {
 	x86_read_percpu(xen_vcpu)->arch.cr2 = cr2;
@@ -624,8 +716,10 @@ static unsigned long xen_read_cr2_direct(void)
 
 static void xen_write_cr4(unsigned long cr4)
 {
-	/* Just ignore cr4 changes; Xen doesn't allow us to do
-	   anything anyway. */
+	cr4 &= ~X86_CR4_PGE;
+	cr4 &= ~X86_CR4_PSE;
+
+	native_write_cr4(cr4);
 }
 
 static unsigned long xen_read_cr3(void)
@@ -638,36 +732,105 @@ static void set_current_cr3(void *v)
 	x86_write_percpu(xen_current_cr3, (unsigned long)v);
 }
 
-static void xen_write_cr3(unsigned long cr3)
+static void __xen_write_cr3(bool kernel, unsigned long cr3)
 {
 	struct mmuext_op *op;
 	struct multicall_space mcs;
-	unsigned long mfn = pfn_to_mfn(PFN_DOWN(cr3));
+	unsigned long mfn;
 
-	BUG_ON(preemptible());
+	if (cr3)
+		mfn = pfn_to_mfn(PFN_DOWN(cr3));
+	else
+		mfn = 0;
 
-	mcs = xen_mc_entry(sizeof(*op));  /* disables interrupts */
+	WARN_ON(mfn == 0 && kernel);
 
-	/* Update while interrupts are disabled, so its atomic with
-	   respect to ipis */
-	x86_write_percpu(xen_cr3, cr3);
+	mcs = __xen_mc_entry(sizeof(*op));
 
 	op = mcs.args;
-	op->cmd = MMUEXT_NEW_BASEPTR;
+	op->cmd = kernel ? MMUEXT_NEW_BASEPTR : MMUEXT_NEW_USER_BASEPTR;
 	op->arg1.mfn = mfn;
 
 	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
 
-	/* Update xen_update_cr3 once the batch has actually
-	   been submitted. */
-	xen_mc_callback(set_current_cr3, (void *)cr3);
+	if (kernel) {
+		x86_write_percpu(xen_cr3, cr3);
+
+		/* Update xen_current_cr3 once the batch has actually
+		   been submitted. */
+		xen_mc_callback(set_current_cr3, (void *)cr3);
+	}
+}
+
+static void xen_write_cr3(unsigned long cr3)
+{
+	BUG_ON(preemptible());
+
+	xen_mc_batch();  /* disables interrupts */
+
+	/* Update while interrupts are disabled, so its atomic with
+	   respect to ipis */
+	x86_write_percpu(xen_cr3, cr3);
+
+	__xen_write_cr3(true, cr3);
+
+#ifdef CONFIG_X86_64
+	{
+		pgd_t *user_pgd = xen_get_user_pgd(__va(cr3));
+		if (user_pgd)
+			__xen_write_cr3(false, __pa(user_pgd));
+		else
+			__xen_write_cr3(false, 0);
+	}
+#endif
 
 	xen_mc_issue(PARAVIRT_LAZY_CPU);  /* interrupts restored */
 }
 
+static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high)
+{
+	int ret;
+
+	ret = 0;
+
+	switch(msr) {
+#ifdef CONFIG_X86_64
+		unsigned which;
+		u64 base;
+
+	case MSR_FS_BASE:		which = SEGBASE_FS; goto set;
+	case MSR_KERNEL_GS_BASE:	which = SEGBASE_GS_USER; goto set;
+	case MSR_GS_BASE:		which = SEGBASE_GS_KERNEL; goto set;
+
+	set:
+		base = ((u64)high << 32) | low;
+		if (HYPERVISOR_set_segment_base(which, base) != 0)
+			ret = -EFAULT;
+		break;
+#endif
+
+	case MSR_STAR:
+	case MSR_CSTAR:
+	case MSR_LSTAR:
+	case MSR_SYSCALL_MASK:
+	case MSR_IA32_SYSENTER_CS:
+	case MSR_IA32_SYSENTER_ESP:
+	case MSR_IA32_SYSENTER_EIP:
+		/* Fast syscall setup is all done in hypercalls, so
+		   these are all ignored.  Stub them out here to stop
+		   Xen console noise. */
+		break;
+
+	default:
+		ret = native_write_msr_safe(msr, low, high);
+	}
+
+	return ret;
+}
+
 /* Early in boot, while setting up the initial pagetable, assume
    everything is pinned. */
-static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn)
+static __init void xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn)
 {
 #ifdef CONFIG_FLATMEM
 	BUG_ON(mem_map);	/* should only be used early */
@@ -677,7 +840,7 @@ static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn)
 
 /* Early release_pte assumes that all pts are pinned, since there's
    only init_mm and anything attached to that is pinned. */
-static void xen_release_pte_init(u32 pfn)
+static void xen_release_pte_init(unsigned long pfn)
 {
 	make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
 }
@@ -693,7 +856,7 @@ static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
 
 /* This needs to make sure the new pte page is pinned iff its being
    attached to a pinned pagetable. */
-static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
+static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned level)
 {
 	struct page *page = pfn_to_page(pfn);
 
@@ -701,8 +864,8 @@ static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
 		SetPagePinned(page);
 
 		if (!PageHighMem(page)) {
-			make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
-			if (level == PT_PTE)
+			make_lowmem_page_readonly(__va(PFN_PHYS((unsigned long)pfn)));
+			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
 				pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
 		} else
 			/* make sure there are no stray mappings of
@@ -711,24 +874,66 @@ static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
 	}
 }
 
-static void xen_alloc_pte(struct mm_struct *mm, u32 pfn)
+static void xen_alloc_pte(struct mm_struct *mm, unsigned long pfn)
 {
 	xen_alloc_ptpage(mm, pfn, PT_PTE);
 }
 
-static void xen_alloc_pmd(struct mm_struct *mm, u32 pfn)
+static void xen_alloc_pmd(struct mm_struct *mm, unsigned long pfn)
 {
 	xen_alloc_ptpage(mm, pfn, PT_PMD);
 }
 
+static int xen_pgd_alloc(struct mm_struct *mm)
+{
+	pgd_t *pgd = mm->pgd;
+	int ret = 0;
+
+	BUG_ON(PagePinned(virt_to_page(pgd)));
+
+#ifdef CONFIG_X86_64
+	{
+		struct page *page = virt_to_page(pgd);
+		pgd_t *user_pgd;
+
+		BUG_ON(page->private != 0);
+
+		ret = -ENOMEM;
+
+		user_pgd = (pgd_t *)__get_free_page(GFP_KERNEL | __GFP_ZERO);
+		page->private = (unsigned long)user_pgd;
+
+		if (user_pgd != NULL) {
+			user_pgd[pgd_index(VSYSCALL_START)] =
+				__pgd(__pa(level3_user_vsyscall) | _PAGE_TABLE);
+			ret = 0;
+		}
+
+		BUG_ON(PagePinned(virt_to_page(xen_get_user_pgd(pgd))));
+	}
+#endif
+
+	return ret;
+}
+
+static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd)
+{
+#ifdef CONFIG_X86_64
+	pgd_t *user_pgd = xen_get_user_pgd(pgd);
+
+	if (user_pgd)
+		free_page((unsigned long)user_pgd);
+#endif
+}
+
 /* This should never happen until we're OK to use struct page */
-static void xen_release_ptpage(u32 pfn, unsigned level)
+static void xen_release_ptpage(unsigned long pfn, unsigned level)
 {
 	struct page *page = pfn_to_page(pfn);
 
 	if (PagePinned(page)) {
 		if (!PageHighMem(page)) {
-			if (level == PT_PTE)
+			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
 				pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
 			make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
 		}
@@ -736,16 +941,28 @@ static void xen_release_ptpage(u32 pfn, unsigned level)
 	}
 }
 
-static void xen_release_pte(u32 pfn)
+static void xen_release_pte(unsigned long pfn)
 {
 	xen_release_ptpage(pfn, PT_PTE);
 }
 
-static void xen_release_pmd(u32 pfn)
+static void xen_release_pmd(unsigned long pfn)
 {
 	xen_release_ptpage(pfn, PT_PMD);
 }
 
+#if PAGETABLE_LEVELS == 4
+static void xen_alloc_pud(struct mm_struct *mm, unsigned long pfn)
+{
+	xen_alloc_ptpage(mm, pfn, PT_PUD);
+}
+
+static void xen_release_pud(unsigned long pfn)
+{
+	xen_release_ptpage(pfn, PT_PUD);
+}
+#endif
+
 #ifdef CONFIG_HIGHPTE
 static void *xen_kmap_atomic_pte(struct page *page, enum km_type type)
 {
@@ -763,6 +980,7 @@ static void *xen_kmap_atomic_pte(struct page *page, enum km_type type)
 }
 #endif
 
+#ifdef CONFIG_X86_32
 static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte)
 {
 	/* If there's an existing pte, then don't allow _PAGE_RW to be set */
@@ -781,71 +999,20 @@ static __init void xen_set_pte_init(pte_t *ptep, pte_t pte)
 
 	xen_set_pte(ptep, pte);
 }
+#endif
 
 static __init void xen_pagetable_setup_start(pgd_t *base)
 {
-	pgd_t *xen_pgd = (pgd_t *)xen_start_info->pt_base;
-	int i;
-
-	/* special set_pte for pagetable initialization */
-	pv_mmu_ops.set_pte = xen_set_pte_init;
-
-	init_mm.pgd = base;
-	/*
-	 * copy top-level of Xen-supplied pagetable into place.  This
-	 * is a stand-in while we copy the pmd pages.
-	 */
-	memcpy(base, xen_pgd, PTRS_PER_PGD * sizeof(pgd_t));
-
-	/*
-	 * For PAE, need to allocate new pmds, rather than
-	 * share Xen's, since Xen doesn't like pmd's being
-	 * shared between address spaces.
-	 */
-	for (i = 0; i < PTRS_PER_PGD; i++) {
-		if (pgd_val_ma(xen_pgd[i]) & _PAGE_PRESENT) {
-			pmd_t *pmd = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE);
-
-			memcpy(pmd, (void *)pgd_page_vaddr(xen_pgd[i]),
-			       PAGE_SIZE);
-
-			make_lowmem_page_readonly(pmd);
-
-			set_pgd(&base[i], __pgd(1 + __pa(pmd)));
-		} else
-			pgd_clear(&base[i]);
-	}
-
-	/* make sure zero_page is mapped RO so we can use it in pagetables */
-	make_lowmem_page_readonly(empty_zero_page);
-	make_lowmem_page_readonly(base);
-	/*
-	 * Switch to new pagetable.  This is done before
-	 * pagetable_init has done anything so that the new pages
-	 * added to the table can be prepared properly for Xen.
-	 */
-	xen_write_cr3(__pa(base));
-
-	/* Unpin initial Xen pagetable */
-	pin_pagetable_pfn(MMUEXT_UNPIN_TABLE,
-			  PFN_DOWN(__pa(xen_start_info->pt_base)));
 }
 
-static __init void setup_shared_info(void)
+void xen_setup_shared_info(void)
 {
 	if (!xen_feature(XENFEAT_auto_translated_physmap)) {
-		unsigned long addr = fix_to_virt(FIX_PARAVIRT_BOOTMAP);
-
-		/*
-		 * Create a mapping for the shared info page.
-		 * Should be set_fixmap(), but shared_info is a machine
-		 * address with no corresponding pseudo-phys address.
-		 */
-		set_pte_mfn(addr,
-			    PFN_DOWN(xen_start_info->shared_info),
-			    PAGE_KERNEL);
-
-		HYPERVISOR_shared_info = (struct shared_info *)addr;
+		set_fixmap(FIX_PARAVIRT_BOOTMAP,
+			   xen_start_info->shared_info);
+
+		HYPERVISOR_shared_info =
+			(struct shared_info *)fix_to_virt(FIX_PARAVIRT_BOOTMAP);
 	} else
 		HYPERVISOR_shared_info =
 			(struct shared_info *)__va(xen_start_info->shared_info);
@@ -854,27 +1021,43 @@ static __init void setup_shared_info(void)
 	/* In UP this is as good a place as any to set up shared info */
 	xen_setup_vcpu_info_placement();
 #endif
+
+	xen_setup_mfn_list_list();
 }
 
 static __init void xen_pagetable_setup_done(pgd_t *base)
 {
+	xen_setup_shared_info();
+}
+
+static __init void xen_post_allocator_init(void)
+{
+	pv_mmu_ops.set_pte = xen_set_pte;
+	pv_mmu_ops.set_pmd = xen_set_pmd;
+	pv_mmu_ops.set_pud = xen_set_pud;
+#if PAGETABLE_LEVELS == 4
+	pv_mmu_ops.set_pgd = xen_set_pgd;
+#endif
+
 	/* This will work as long as patching hasn't happened yet
 	   (which it hasn't) */
 	pv_mmu_ops.alloc_pte = xen_alloc_pte;
 	pv_mmu_ops.alloc_pmd = xen_alloc_pmd;
 	pv_mmu_ops.release_pte = xen_release_pte;
 	pv_mmu_ops.release_pmd = xen_release_pmd;
-	pv_mmu_ops.set_pte = xen_set_pte;
-
-	setup_shared_info();
+#if PAGETABLE_LEVELS == 4
+	pv_mmu_ops.alloc_pud = xen_alloc_pud;
+	pv_mmu_ops.release_pud = xen_release_pud;
+#endif
 
-	/* Actually pin the pagetable down, but we can't set PG_pinned
-	   yet because the page structures don't exist yet. */
-	pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(base)));
+#ifdef CONFIG_X86_64
+	SetPagePinned(virt_to_page(level3_user_vsyscall));
+#endif
+	xen_mark_init_mm_pinned();
 }
 
 /* This is called once we have the cpu_possible_map */
-void __init xen_setup_vcpu_info_placement(void)
+void xen_setup_vcpu_info_placement(void)
 {
 	int cpu;
 
@@ -947,6 +1130,49 @@ static unsigned xen_patch(u8 type, u16 clobbers, void *insnbuf,
 	return ret;
 }
 
+static void xen_set_fixmap(unsigned idx, unsigned long phys, pgprot_t prot)
+{
+	pte_t pte;
+
+	phys >>= PAGE_SHIFT;
+
+	switch (idx) {
+	case FIX_BTMAP_END ... FIX_BTMAP_BEGIN:
+#ifdef CONFIG_X86_F00F_BUG
+	case FIX_F00F_IDT:
+#endif
+#ifdef CONFIG_X86_32
+	case FIX_WP_TEST:
+	case FIX_VDSO:
+# ifdef CONFIG_HIGHMEM
+	case FIX_KMAP_BEGIN ... FIX_KMAP_END:
+# endif
+#else
+	case VSYSCALL_LAST_PAGE ... VSYSCALL_FIRST_PAGE:
+#endif
+#ifdef CONFIG_X86_LOCAL_APIC
+	case FIX_APIC_BASE:	/* maps dummy local APIC */
+#endif
+		pte = pfn_pte(phys, prot);
+		break;
+
+	default:
+		pte = mfn_pte(phys, prot);
+		break;
+	}
+
+	__native_set_fixmap(idx, pte);
+
+#ifdef CONFIG_X86_64
+	/* Replicate changes to map the vsyscall page into the user
+	   pagetable vsyscall mapping. */
+	if (idx >= VSYSCALL_LAST_PAGE && idx <= VSYSCALL_FIRST_PAGE) {
+		unsigned long vaddr = __fix_to_virt(idx);
+		set_pte_vaddr_pud(level3_user_vsyscall, vaddr, pte);
+	}
+#endif
+}
+
 static const struct pv_info xen_info __initdata = {
 	.paravirt_enabled = 1,
 	.shared_kernel_pmd = 0,
@@ -960,7 +1186,7 @@ static const struct pv_init_ops xen_init_ops __initdata = {
 	.banner = xen_banner,
 	.memory_setup = xen_memory_setup,
 	.arch_setup = xen_arch_setup,
-	.post_allocator_init = xen_mark_init_mm_pinned,
+	.post_allocator_init = xen_post_allocator_init,
 };
 
 static const struct pv_time_ops xen_time_ops __initdata = {
@@ -968,7 +1194,7 @@ static const struct pv_time_ops xen_time_ops __initdata = {
 
 	.set_wallclock = xen_set_wallclock,
 	.get_wallclock = xen_get_wallclock,
-	.get_cpu_khz = xen_cpu_khz,
+	.get_tsc_khz = xen_tsc_khz,
 	.sched_clock = xen_sched_clock,
 };
 
@@ -978,10 +1204,10 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = {
 	.set_debugreg = xen_set_debugreg,
 	.get_debugreg = xen_get_debugreg,
 
-	.clts = native_clts,
+	.clts = xen_clts,
 
 	.read_cr0 = native_read_cr0,
-	.write_cr0 = native_write_cr0,
+	.write_cr0 = xen_write_cr0,
 
 	.read_cr4 = native_read_cr4,
 	.read_cr4_safe = native_read_cr4_safe,
@@ -990,18 +1216,28 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = {
 	.wbinvd = native_wbinvd,
 
 	.read_msr = native_read_msr_safe,
-	.write_msr = native_write_msr_safe,
+	.write_msr = xen_write_msr_safe,
 	.read_tsc = native_read_tsc,
 	.read_pmc = native_read_pmc,
 
 	.iret = xen_iret,
-	.irq_enable_syscall_ret = xen_sysexit,
+	.irq_enable_sysexit = xen_sysexit,
+#ifdef CONFIG_X86_64
+	.usergs_sysret32 = xen_sysret32,
+	.usergs_sysret64 = xen_sysret64,
+#endif
 
 	.load_tr_desc = paravirt_nop,
 	.set_ldt = xen_set_ldt,
 	.load_gdt = xen_load_gdt,
 	.load_idt = xen_load_idt,
 	.load_tls = xen_load_tls,
+#ifdef CONFIG_X86_64
+	.load_gs_index = xen_load_gs_index,
+#endif
+
+	.alloc_ldt = xen_alloc_ldt,
+	.free_ldt = xen_free_ldt,
 
 	.store_gdt = native_store_gdt,
 	.store_idt = native_store_idt,
@@ -1015,27 +1251,17 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = {
 	.set_iopl_mask = xen_set_iopl_mask,
 	.io_delay = xen_io_delay,
 
+	/* Xen takes care of %gs when switching to usermode for us */
+	.swapgs = paravirt_nop,
+
 	.lazy_mode = {
 		.enter = paravirt_enter_lazy_cpu,
 		.leave = xen_leave_lazy,
 	},
 };
 
-static const struct pv_irq_ops xen_irq_ops __initdata = {
-	.init_IRQ = xen_init_IRQ,
-	.save_fl = xen_save_fl,
-	.restore_fl = xen_restore_fl,
-	.irq_disable = xen_irq_disable,
-	.irq_enable = xen_irq_enable,
-	.safe_halt = xen_safe_halt,
-	.halt = xen_halt,
-};
-
 static const struct pv_apic_ops xen_apic_ops __initdata = {
 #ifdef CONFIG_X86_LOCAL_APIC
-	.apic_write = xen_apic_write,
-	.apic_write_atomic = xen_apic_write,
-	.apic_read = xen_apic_read,
 	.setup_boot_clock = paravirt_nop,
 	.setup_secondary_clock = paravirt_nop,
 	.startup_ipi_hook = paravirt_nop,
@@ -1060,6 +1286,9 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = {
 	.pte_update = paravirt_nop,
 	.pte_update_defer = paravirt_nop,
 
+	.pgd_alloc = xen_pgd_alloc,
+	.pgd_free = xen_pgd_free,
+
 	.alloc_pte = xen_alloc_pte_init,
 	.release_pte = xen_release_pte_init,
 	.alloc_pmd = xen_alloc_pte_init,
@@ -1070,25 +1299,44 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = {
 	.kmap_atomic_pte = xen_kmap_atomic_pte,
 #endif
 
-	.set_pte = NULL,	/* see xen_pagetable_setup_* */
+#ifdef CONFIG_X86_64
+	.set_pte = xen_set_pte,
+#else
+	.set_pte = xen_set_pte_init,
+#endif
 	.set_pte_at = xen_set_pte_at,
-	.set_pmd = xen_set_pmd,
+	.set_pmd = xen_set_pmd_hyper,
+
+	.ptep_modify_prot_start = __ptep_modify_prot_start,
+	.ptep_modify_prot_commit = __ptep_modify_prot_commit,
 
 	.pte_val = xen_pte_val,
+	.pte_flags = native_pte_flags,
 	.pgd_val = xen_pgd_val,
 
 	.make_pte = xen_make_pte,
 	.make_pgd = xen_make_pgd,
 
+#ifdef CONFIG_X86_PAE
 	.set_pte_atomic = xen_set_pte_atomic,
 	.set_pte_present = xen_set_pte_at,
-	.set_pud = xen_set_pud,
 	.pte_clear = xen_pte_clear,
 	.pmd_clear = xen_pmd_clear,
+#endif	/* CONFIG_X86_PAE */
+	.set_pud = xen_set_pud_hyper,
 
 	.make_pmd = xen_make_pmd,
 	.pmd_val = xen_pmd_val,
 
+#if PAGETABLE_LEVELS == 4
+	.pud_val = xen_pud_val,
+	.make_pud = xen_make_pud,
+	.set_pgd = xen_set_pgd_hyper,
+
+	.alloc_pud = xen_alloc_pte_init,
+	.release_pud = xen_release_pte_init,
+#endif	/* PAGETABLE_LEVELS == 4 */
+
 	.activate_mm = xen_activate_mm,
 	.dup_mmap = xen_dup_mmap,
 	.exit_mmap = xen_exit_mmap,
@@ -1097,28 +1345,19 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = {
 		.enter = paravirt_enter_lazy_mmu,
 		.leave = xen_leave_lazy,
 	},
-};
 
-#ifdef CONFIG_SMP
-static const struct smp_ops xen_smp_ops __initdata = {
-	.smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu,
-	.smp_prepare_cpus = xen_smp_prepare_cpus,
-	.cpu_up = xen_cpu_up,
-	.smp_cpus_done = xen_smp_cpus_done,
-
-	.smp_send_stop = xen_smp_send_stop,
-	.smp_send_reschedule = xen_smp_send_reschedule,
-	.smp_call_function_mask = xen_smp_call_function_mask,
+	.set_fixmap = xen_set_fixmap,
 };
-#endif	/* CONFIG_SMP */
 
 static void xen_reboot(int reason)
 {
+	struct sched_shutdown r = { .reason = reason };
+
 #ifdef CONFIG_SMP
 	smp_send_stop();
 #endif
 
-	if (HYPERVISOR_sched_op(SCHEDOP_shutdown, reason))
+	if (HYPERVISOR_sched_op(SCHEDOP_shutdown, &r))
 		BUG();
 }
 
@@ -1154,15 +1393,219 @@ static const struct machine_ops __initdata xen_machine_ops = {
 
 static void __init xen_reserve_top(void)
 {
+#ifdef CONFIG_X86_32
 	unsigned long top = HYPERVISOR_VIRT_START;
 	struct xen_platform_parameters pp;
 
 	if (HYPERVISOR_xen_version(XENVER_platform_parameters, &pp) == 0)
 		top = pp.virt_start;
 
-	reserve_top_address(-top + 2 * PAGE_SIZE);
+	reserve_top_address(-top);
+#endif	/* CONFIG_X86_32 */
+}
+
+/*
+ * Like __va(), but returns address in the kernel mapping (which is
+ * all we have until the physical memory mapping has been set up.
+ */
+static void *__ka(phys_addr_t paddr)
+{
+#ifdef CONFIG_X86_64
+	return (void *)(paddr + __START_KERNEL_map);
+#else
+	return __va(paddr);
+#endif
+}
+
+/* Convert a machine address to physical address */
+static unsigned long m2p(phys_addr_t maddr)
+{
+	phys_addr_t paddr;
+
+	maddr &= PTE_PFN_MASK;
+	paddr = mfn_to_pfn(maddr >> PAGE_SHIFT) << PAGE_SHIFT;
+
+	return paddr;
 }
 
+/* Convert a machine address to kernel virtual */
+static void *m2v(phys_addr_t maddr)
+{
+	return __ka(m2p(maddr));
+}
+
+static void set_page_prot(void *addr, pgprot_t prot)
+{
+	unsigned long pfn = __pa(addr) >> PAGE_SHIFT;
+	pte_t pte = pfn_pte(pfn, prot);
+
+	if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, 0))
+		BUG();
+}
+
+static __init void xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn)
+{
+	unsigned pmdidx, pteidx;
+	unsigned ident_pte;
+	unsigned long pfn;
+
+	ident_pte = 0;
+	pfn = 0;
+	for(pmdidx = 0; pmdidx < PTRS_PER_PMD && pfn < max_pfn; pmdidx++) {
+		pte_t *pte_page;
+
+		/* Reuse or allocate a page of ptes */
+		if (pmd_present(pmd[pmdidx]))
+			pte_page = m2v(pmd[pmdidx].pmd);
+		else {
+			/* Check for free pte pages */
+			if (ident_pte == ARRAY_SIZE(level1_ident_pgt))
+				break;
+
+			pte_page = &level1_ident_pgt[ident_pte];
+			ident_pte += PTRS_PER_PTE;
+
+			pmd[pmdidx] = __pmd(__pa(pte_page) | _PAGE_TABLE);
+		}
+
+		/* Install mappings */
+		for(pteidx = 0; pteidx < PTRS_PER_PTE; pteidx++, pfn++) {
+			pte_t pte;
+
+			if (pfn > max_pfn_mapped)
+				max_pfn_mapped = pfn;
+
+			if (!pte_none(pte_page[pteidx]))
+				continue;
+
+			pte = pfn_pte(pfn, PAGE_KERNEL_EXEC);
+			pte_page[pteidx] = pte;
+		}
+	}
+
+	for(pteidx = 0; pteidx < ident_pte; pteidx += PTRS_PER_PTE)
+		set_page_prot(&level1_ident_pgt[pteidx], PAGE_KERNEL_RO);
+
+	set_page_prot(pmd, PAGE_KERNEL_RO);
+}
+
+#ifdef CONFIG_X86_64
+static void convert_pfn_mfn(void *v)
+{
+	pte_t *pte = v;
+	int i;
+
+	/* All levels are converted the same way, so just treat them
+	   as ptes. */
+	for(i = 0; i < PTRS_PER_PTE; i++)
+		pte[i] = xen_make_pte(pte[i].pte);
+}
+
+/*
+ * Set up the inital kernel pagetable.
+ *
+ * We can construct this by grafting the Xen provided pagetable into
+ * head_64.S's preconstructed pagetables.  We copy the Xen L2's into
+ * level2_ident_pgt, level2_kernel_pgt and level2_fixmap_pgt.  This
+ * means that only the kernel has a physical mapping to start with -
+ * but that's enough to get __va working.  We need to fill in the rest
+ * of the physical mapping once some sort of allocator has been set
+ * up.
+ */
+static __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
+{
+	pud_t *l3;
+	pmd_t *l2;
+
+	/* Zap identity mapping */
+	init_level4_pgt[0] = __pgd(0);
+
+	/* Pre-constructed entries are in pfn, so convert to mfn */
+	convert_pfn_mfn(init_level4_pgt);
+	convert_pfn_mfn(level3_ident_pgt);
+	convert_pfn_mfn(level3_kernel_pgt);
+
+	l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd);
+	l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud);
+
+	memcpy(level2_ident_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
+	memcpy(level2_kernel_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
+
+	l3 = m2v(pgd[pgd_index(__START_KERNEL_map + PMD_SIZE)].pgd);
+	l2 = m2v(l3[pud_index(__START_KERNEL_map + PMD_SIZE)].pud);
+	memcpy(level2_fixmap_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
+
+	/* Set up identity map */
+	xen_map_identity_early(level2_ident_pgt, max_pfn);
+
+	/* Make pagetable pieces RO */
+	set_page_prot(init_level4_pgt, PAGE_KERNEL_RO);
+	set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO);
+	set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO);
+	set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO);
+	set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO);
+	set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO);
+
+	/* Pin down new L4 */
+	pin_pagetable_pfn(MMUEXT_PIN_L4_TABLE,
+			  PFN_DOWN(__pa_symbol(init_level4_pgt)));
+
+	/* Unpin Xen-provided one */
+	pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
+
+	/* Switch over */
+	pgd = init_level4_pgt;
+
+	/*
+	 * At this stage there can be no user pgd, and no page
+	 * structure to attach it to, so make sure we just set kernel
+	 * pgd.
+	 */
+	xen_mc_batch();
+	__xen_write_cr3(true, __pa(pgd));
+	xen_mc_issue(PARAVIRT_LAZY_CPU);
+
+	reserve_early(__pa(xen_start_info->pt_base),
+		      __pa(xen_start_info->pt_base +
+			   xen_start_info->nr_pt_frames * PAGE_SIZE),
+		      "XEN PAGETABLES");
+
+	return pgd;
+}
+#else	/* !CONFIG_X86_64 */
+static pmd_t level2_kernel_pgt[PTRS_PER_PMD] __page_aligned_bss;
+
+static __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
+{
+	pmd_t *kernel_pmd;
+
+	init_pg_tables_start = __pa(pgd);
+	init_pg_tables_end = __pa(pgd) + xen_start_info->nr_pt_frames*PAGE_SIZE;
+	max_pfn_mapped = PFN_DOWN(init_pg_tables_end + 512*1024);
+
+	kernel_pmd = m2v(pgd[KERNEL_PGD_BOUNDARY].pgd);
+	memcpy(level2_kernel_pgt, kernel_pmd, sizeof(pmd_t) * PTRS_PER_PMD);
+
+	xen_map_identity_early(level2_kernel_pgt, max_pfn);
+
+	memcpy(swapper_pg_dir, pgd, sizeof(pgd_t) * PTRS_PER_PGD);
+	set_pgd(&swapper_pg_dir[KERNEL_PGD_BOUNDARY],
+			__pgd(__pa(level2_kernel_pgt) | _PAGE_PRESENT));
+
+	set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO);
+	set_page_prot(swapper_pg_dir, PAGE_KERNEL_RO);
+	set_page_prot(empty_zero_page, PAGE_KERNEL_RO);
+
+	pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
+
+	xen_write_cr3(__pa(swapper_pg_dir));
+
+	pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(swapper_pg_dir)));
+
+	return swapper_pg_dir;
+}
+#endif	/* CONFIG_X86_64 */
+
 /* First C function to be called on Xen boot */
 asmlinkage void __init xen_start_kernel(void)
 {
@@ -1171,70 +1614,99 @@ asmlinkage void __init xen_start_kernel(void)
 	if (!xen_start_info)
 		return;
 
+	xen_domain_type = XEN_PV_DOMAIN;
+
 	BUG_ON(memcmp(xen_start_info->magic, "xen-3", 5) != 0);
 
+	xen_setup_features();
+
 	/* Install Xen paravirt ops */
 	pv_info = xen_info;
 	pv_init_ops = xen_init_ops;
 	pv_time_ops = xen_time_ops;
 	pv_cpu_ops = xen_cpu_ops;
-	pv_irq_ops = xen_irq_ops;
 	pv_apic_ops = xen_apic_ops;
 	pv_mmu_ops = xen_mmu_ops;
 
+	xen_init_irq_ops();
+
+#ifdef CONFIG_X86_LOCAL_APIC
+	/*
+	 * set up the basic apic ops.
+	 */
+	apic_ops = &xen_basic_apic_ops;
+#endif
+
+	if (xen_feature(XENFEAT_mmu_pt_update_preserve_ad)) {
+		pv_mmu_ops.ptep_modify_prot_start = xen_ptep_modify_prot_start;
+		pv_mmu_ops.ptep_modify_prot_commit = xen_ptep_modify_prot_commit;
+	}
+
 	machine_ops = xen_machine_ops;
 
-#ifdef CONFIG_SMP
-	smp_ops = xen_smp_ops;
+#ifdef CONFIG_X86_64
+	/* Disable until direct per-cpu data access. */
+	have_vcpu_info_placement = 0;
+	x86_64_init_pda();
 #endif
 
-	xen_setup_features();
+	xen_smp_init();
 
 	/* Get mfn list */
 	if (!xen_feature(XENFEAT_auto_translated_physmap))
-		phys_to_machine_mapping = (unsigned long *)xen_start_info->mfn_list;
+		xen_build_dynamic_phys_to_machine();
 
 	pgd = (pgd_t *)xen_start_info->pt_base;
 
-	init_pg_tables_end = __pa(pgd) + xen_start_info->nr_pt_frames*PAGE_SIZE;
-
-	init_mm.pgd = pgd; /* use the Xen pagetables to start */
-
-	/* keep using Xen gdt for now; no urgent need to change it */
-
-	x86_write_percpu(xen_cr3, __pa(pgd));
-	x86_write_percpu(xen_current_cr3, __pa(pgd));
+	/* Prevent unwanted bits from being set in PTEs. */
+	__supported_pte_mask &= ~_PAGE_GLOBAL;
+	if (!xen_initial_domain())
+		__supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD);
 
 	/* Don't do the full vcpu_info placement stuff until we have a
 	   possible map and a non-dummy shared_info. */
 	per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0];
 
+	xen_raw_console_write("mapping kernel into physical memory\n");
+	pgd = xen_setup_kernel_pagetable(pgd, xen_start_info->nr_pages);
+
+	init_mm.pgd = pgd;
+
+	/* keep using Xen gdt for now; no urgent need to change it */
+
 	pv_info.kernel_rpl = 1;
 	if (xen_feature(XENFEAT_supervisor_mode_kernel))
 		pv_info.kernel_rpl = 0;
 
-	/* Prevent unwanted bits from being set in PTEs. */
-	__supported_pte_mask &= ~_PAGE_GLOBAL;
-	if (!is_initial_xendomain())
-		__supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD);
-
 	/* set the limit of our address space */
 	xen_reserve_top();
 
+#ifdef CONFIG_X86_32
 	/* set up basic CPUID stuff */
 	cpu_detect(&new_cpu_data);
 	new_cpu_data.hard_math = 1;
 	new_cpu_data.x86_capability[0] = cpuid_edx(1);
+#endif
 
 	/* Poke various useful things into boot_params */
 	boot_params.hdr.type_of_loader = (9 << 4) | 0;
 	boot_params.hdr.ramdisk_image = xen_start_info->mod_start
 		? __pa(xen_start_info->mod_start) : 0;
 	boot_params.hdr.ramdisk_size = xen_start_info->mod_len;
+	boot_params.hdr.cmd_line_ptr = __pa(xen_start_info->cmd_line);
 
-	if (!is_initial_xendomain())
+	if (!xen_initial_domain()) {
+		add_preferred_console("xenboot", 0, NULL);
+		add_preferred_console("tty", 0, NULL);
 		add_preferred_console("hvc", 0, NULL);
+	}
+
+	xen_raw_console_write("about to get started...\n");
 
 	/* Start the world */
-	start_kernel();
+#ifdef CONFIG_X86_32
+	i386_start_kernel();
+#else
+	x86_64_start_reservations((char *)__pa_symbol(&boot_params));
+#endif
 }
diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c
new file mode 100644
index 00000000000..28b85ab8422
--- /dev/null
+++ b/arch/x86/xen/irq.c
@@ -0,0 +1,143 @@
+#include <linux/hardirq.h>
+
+#include <xen/interface/xen.h>
+#include <xen/interface/sched.h>
+#include <xen/interface/vcpu.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/hypervisor.h>
+
+#include "xen-ops.h"
+
+/*
+ * Force a proper event-channel callback from Xen after clearing the
+ * callback mask. We do this in a very simple manner, by making a call
+ * down into Xen. The pending flag will be checked by Xen on return.
+ */
+void xen_force_evtchn_callback(void)
+{
+	(void)HYPERVISOR_xen_version(0, NULL);
+}
+
+static void __init __xen_init_IRQ(void)
+{
+#ifdef CONFIG_X86_64
+	int i;
+
+	/* Create identity vector->irq map */
+	for(i = 0; i < NR_VECTORS; i++) {
+		int cpu;
+
+		for_each_possible_cpu(cpu)
+			per_cpu(vector_irq, cpu)[i] = i;
+	}
+#endif	/* CONFIG_X86_64 */
+
+	xen_init_IRQ();
+}
+
+static unsigned long xen_save_fl(void)
+{
+	struct vcpu_info *vcpu;
+	unsigned long flags;
+
+	vcpu = x86_read_percpu(xen_vcpu);
+
+	/* flag has opposite sense of mask */
+	flags = !vcpu->evtchn_upcall_mask;
+
+	/* convert to IF type flag
+	   -0 -> 0x00000000
+	   -1 -> 0xffffffff
+	*/
+	return (-flags) & X86_EFLAGS_IF;
+}
+
+static void xen_restore_fl(unsigned long flags)
+{
+	struct vcpu_info *vcpu;
+
+	/* convert from IF type flag */
+	flags = !(flags & X86_EFLAGS_IF);
+
+	/* There's a one instruction preempt window here.  We need to
+	   make sure we're don't switch CPUs between getting the vcpu
+	   pointer and updating the mask. */
+	preempt_disable();
+	vcpu = x86_read_percpu(xen_vcpu);
+	vcpu->evtchn_upcall_mask = flags;
+	preempt_enable_no_resched();
+
+	/* Doesn't matter if we get preempted here, because any
+	   pending event will get dealt with anyway. */
+
+	if (flags == 0) {
+		preempt_check_resched();
+		barrier(); /* unmask then check (avoid races) */
+		if (unlikely(vcpu->evtchn_upcall_pending))
+			xen_force_evtchn_callback();
+	}
+}
+
+static void xen_irq_disable(void)
+{
+	/* There's a one instruction preempt window here.  We need to
+	   make sure we're don't switch CPUs between getting the vcpu
+	   pointer and updating the mask. */
+	preempt_disable();
+	x86_read_percpu(xen_vcpu)->evtchn_upcall_mask = 1;
+	preempt_enable_no_resched();
+}
+
+static void xen_irq_enable(void)
+{
+	struct vcpu_info *vcpu;
+
+	/* We don't need to worry about being preempted here, since
+	   either a) interrupts are disabled, so no preemption, or b)
+	   the caller is confused and is trying to re-enable interrupts
+	   on an indeterminate processor. */
+
+	vcpu = x86_read_percpu(xen_vcpu);
+	vcpu->evtchn_upcall_mask = 0;
+
+	/* Doesn't matter if we get preempted here, because any
+	   pending event will get dealt with anyway. */
+
+	barrier(); /* unmask then check (avoid races) */
+	if (unlikely(vcpu->evtchn_upcall_pending))
+		xen_force_evtchn_callback();
+}
+
+static void xen_safe_halt(void)
+{
+	/* Blocking includes an implicit local_irq_enable(). */
+	if (HYPERVISOR_sched_op(SCHEDOP_block, NULL) != 0)
+		BUG();
+}
+
+static void xen_halt(void)
+{
+	if (irqs_disabled())
+		HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
+	else
+		xen_safe_halt();
+}
+
+static const struct pv_irq_ops xen_irq_ops __initdata = {
+	.init_IRQ = __xen_init_IRQ,
+	.save_fl = xen_save_fl,
+	.restore_fl = xen_restore_fl,
+	.irq_disable = xen_irq_disable,
+	.irq_enable = xen_irq_enable,
+	.safe_halt = xen_safe_halt,
+	.halt = xen_halt,
+#ifdef CONFIG_X86_64
+	.adjust_exception_frame = xen_adjust_exception_frame,
+#endif
+};
+
+void __init xen_init_irq_ops()
+{
+	pv_irq_ops = xen_irq_ops;
+}
diff --git a/arch/x86/xen/manage.c b/arch/x86/xen/manage.c
deleted file mode 100644
index aa7af9e6abc..00000000000
--- a/arch/x86/xen/manage.c
+++ /dev/null
@@ -1,143 +0,0 @@
-/*
- * Handle extern requests for shutdown, reboot and sysrq
- */
-#include <linux/kernel.h>
-#include <linux/err.h>
-#include <linux/reboot.h>
-#include <linux/sysrq.h>
-
-#include <xen/xenbus.h>
-
-#define SHUTDOWN_INVALID  -1
-#define SHUTDOWN_POWEROFF  0
-#define SHUTDOWN_SUSPEND   2
-/* Code 3 is SHUTDOWN_CRASH, which we don't use because the domain can only
- * report a crash, not be instructed to crash!
- * HALT is the same as POWEROFF, as far as we're concerned.  The tools use
- * the distinction when we return the reason code to them.
- */
-#define SHUTDOWN_HALT      4
-
-/* Ignore multiple shutdown requests. */
-static int shutting_down = SHUTDOWN_INVALID;
-
-static void shutdown_handler(struct xenbus_watch *watch,
-			     const char **vec, unsigned int len)
-{
-	char *str;
-	struct xenbus_transaction xbt;
-	int err;
-
-	if (shutting_down != SHUTDOWN_INVALID)
-		return;
-
- again:
-	err = xenbus_transaction_start(&xbt);
-	if (err)
-		return;
-
-	str = (char *)xenbus_read(xbt, "control", "shutdown", NULL);
-	/* Ignore read errors and empty reads. */
-	if (XENBUS_IS_ERR_READ(str)) {
-		xenbus_transaction_end(xbt, 1);
-		return;
-	}
-
-	xenbus_write(xbt, "control", "shutdown", "");
-
-	err = xenbus_transaction_end(xbt, 0);
-	if (err == -EAGAIN) {
-		kfree(str);
-		goto again;
-	}
-
-	if (strcmp(str, "poweroff") == 0 ||
-	    strcmp(str, "halt") == 0)
-		orderly_poweroff(false);
-	else if (strcmp(str, "reboot") == 0)
-		ctrl_alt_del();
-	else {
-		printk(KERN_INFO "Ignoring shutdown request: %s\n", str);
-		shutting_down = SHUTDOWN_INVALID;
-	}
-
-	kfree(str);
-}
-
-static void sysrq_handler(struct xenbus_watch *watch, const char **vec,
-			  unsigned int len)
-{
-	char sysrq_key = '\0';
-	struct xenbus_transaction xbt;
-	int err;
-
- again:
-	err = xenbus_transaction_start(&xbt);
-	if (err)
-		return;
-	if (!xenbus_scanf(xbt, "control", "sysrq", "%c", &sysrq_key)) {
-		printk(KERN_ERR "Unable to read sysrq code in "
-		       "control/sysrq\n");
-		xenbus_transaction_end(xbt, 1);
-		return;
-	}
-
-	if (sysrq_key != '\0')
-		xenbus_printf(xbt, "control", "sysrq", "%c", '\0');
-
-	err = xenbus_transaction_end(xbt, 0);
-	if (err == -EAGAIN)
-		goto again;
-
-	if (sysrq_key != '\0')
-		handle_sysrq(sysrq_key, NULL);
-}
-
-static struct xenbus_watch shutdown_watch = {
-	.node = "control/shutdown",
-	.callback = shutdown_handler
-};
-
-static struct xenbus_watch sysrq_watch = {
-	.node = "control/sysrq",
-	.callback = sysrq_handler
-};
-
-static int setup_shutdown_watcher(void)
-{
-	int err;
-
-	err = register_xenbus_watch(&shutdown_watch);
-	if (err) {
-		printk(KERN_ERR "Failed to set shutdown watcher\n");
-		return err;
-	}
-
-	err = register_xenbus_watch(&sysrq_watch);
-	if (err) {
-		printk(KERN_ERR "Failed to set sysrq watcher\n");
-		return err;
-	}
-
-	return 0;
-}
-
-static int shutdown_event(struct notifier_block *notifier,
-			  unsigned long event,
-			  void *data)
-{
-	setup_shutdown_watcher();
-	return NOTIFY_DONE;
-}
-
-static int __init setup_shutdown_event(void)
-{
-	static struct notifier_block xenstore_notifier = {
-		.notifier_call = shutdown_event
-	};
-	register_xenstore_notifier(&xenstore_notifier);
-
-	return 0;
-}
-
-subsys_initcall(setup_shutdown_event);
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index df40bf74ea7..ae173f6edd8 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -40,12 +40,15 @@
  */
 #include <linux/sched.h>
 #include <linux/highmem.h>
+#include <linux/debugfs.h>
 #include <linux/bug.h>
 
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
+#include <asm/fixmap.h>
 #include <asm/mmu_context.h>
 #include <asm/paravirt.h>
+#include <asm/linkage.h>
 
 #include <asm/xen/hypercall.h>
 #include <asm/xen/hypervisor.h>
@@ -55,16 +58,200 @@
 
 #include "multicalls.h"
 #include "mmu.h"
+#include "debugfs.h"
 
-xmaddr_t arbitrary_virt_to_machine(unsigned long address)
+#define MMU_UPDATE_HISTO	30
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct {
+	u32 pgd_update;
+	u32 pgd_update_pinned;
+	u32 pgd_update_batched;
+
+	u32 pud_update;
+	u32 pud_update_pinned;
+	u32 pud_update_batched;
+
+	u32 pmd_update;
+	u32 pmd_update_pinned;
+	u32 pmd_update_batched;
+
+	u32 pte_update;
+	u32 pte_update_pinned;
+	u32 pte_update_batched;
+
+	u32 mmu_update;
+	u32 mmu_update_extended;
+	u32 mmu_update_histo[MMU_UPDATE_HISTO];
+
+	u32 prot_commit;
+	u32 prot_commit_batched;
+
+	u32 set_pte_at;
+	u32 set_pte_at_batched;
+	u32 set_pte_at_pinned;
+	u32 set_pte_at_current;
+	u32 set_pte_at_kernel;
+} mmu_stats;
+
+static u8 zero_stats;
+
+static inline void check_zero(void)
+{
+	if (unlikely(zero_stats)) {
+		memset(&mmu_stats, 0, sizeof(mmu_stats));
+		zero_stats = 0;
+	}
+}
+
+#define ADD_STATS(elem, val)			\
+	do { check_zero(); mmu_stats.elem += (val); } while(0)
+
+#else  /* !CONFIG_XEN_DEBUG_FS */
+
+#define ADD_STATS(elem, val)	do { (void)(val); } while(0)
+
+#endif /* CONFIG_XEN_DEBUG_FS */
+
+/*
+ * Just beyond the highest usermode address.  STACK_TOP_MAX has a
+ * redzone above it, so round it up to a PGD boundary.
+ */
+#define USER_LIMIT	((STACK_TOP_MAX + PGDIR_SIZE - 1) & PGDIR_MASK)
+
+
+#define P2M_ENTRIES_PER_PAGE	(PAGE_SIZE / sizeof(unsigned long))
+#define TOP_ENTRIES		(MAX_DOMAIN_PAGES / P2M_ENTRIES_PER_PAGE)
+
+/* Placeholder for holes in the address space */
+static unsigned long p2m_missing[P2M_ENTRIES_PER_PAGE] __page_aligned_data =
+		{ [ 0 ... P2M_ENTRIES_PER_PAGE-1 ] = ~0UL };
+
+ /* Array of pointers to pages containing p2m entries */
+static unsigned long *p2m_top[TOP_ENTRIES] __page_aligned_data =
+		{ [ 0 ... TOP_ENTRIES - 1] = &p2m_missing[0] };
+
+/* Arrays of p2m arrays expressed in mfns used for save/restore */
+static unsigned long p2m_top_mfn[TOP_ENTRIES] __page_aligned_bss;
+
+static unsigned long p2m_top_mfn_list[TOP_ENTRIES / P2M_ENTRIES_PER_PAGE]
+	__page_aligned_bss;
+
+static inline unsigned p2m_top_index(unsigned long pfn)
+{
+	BUG_ON(pfn >= MAX_DOMAIN_PAGES);
+	return pfn / P2M_ENTRIES_PER_PAGE;
+}
+
+static inline unsigned p2m_index(unsigned long pfn)
+{
+	return pfn % P2M_ENTRIES_PER_PAGE;
+}
+
+/* Build the parallel p2m_top_mfn structures */
+void xen_setup_mfn_list_list(void)
 {
+	unsigned pfn, idx;
+
+	for(pfn = 0; pfn < MAX_DOMAIN_PAGES; pfn += P2M_ENTRIES_PER_PAGE) {
+		unsigned topidx = p2m_top_index(pfn);
+
+		p2m_top_mfn[topidx] = virt_to_mfn(p2m_top[topidx]);
+	}
+
+	for(idx = 0; idx < ARRAY_SIZE(p2m_top_mfn_list); idx++) {
+		unsigned topidx = idx * P2M_ENTRIES_PER_PAGE;
+		p2m_top_mfn_list[idx] = virt_to_mfn(&p2m_top_mfn[topidx]);
+	}
+
+	BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
+
+	HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
+		virt_to_mfn(p2m_top_mfn_list);
+	HYPERVISOR_shared_info->arch.max_pfn = xen_start_info->nr_pages;
+}
+
+/* Set up p2m_top to point to the domain-builder provided p2m pages */
+void __init xen_build_dynamic_phys_to_machine(void)
+{
+	unsigned long *mfn_list = (unsigned long *)xen_start_info->mfn_list;
+	unsigned long max_pfn = min(MAX_DOMAIN_PAGES, xen_start_info->nr_pages);
+	unsigned pfn;
+
+	for(pfn = 0; pfn < max_pfn; pfn += P2M_ENTRIES_PER_PAGE) {
+		unsigned topidx = p2m_top_index(pfn);
+
+		p2m_top[topidx] = &mfn_list[pfn];
+	}
+}
+
+unsigned long get_phys_to_machine(unsigned long pfn)
+{
+	unsigned topidx, idx;
+
+	if (unlikely(pfn >= MAX_DOMAIN_PAGES))
+		return INVALID_P2M_ENTRY;
+
+	topidx = p2m_top_index(pfn);
+	idx = p2m_index(pfn);
+	return p2m_top[topidx][idx];
+}
+EXPORT_SYMBOL_GPL(get_phys_to_machine);
+
+static void alloc_p2m(unsigned long **pp, unsigned long *mfnp)
+{
+	unsigned long *p;
+	unsigned i;
+
+	p = (void *)__get_free_page(GFP_KERNEL | __GFP_NOFAIL);
+	BUG_ON(p == NULL);
+
+	for(i = 0; i < P2M_ENTRIES_PER_PAGE; i++)
+		p[i] = INVALID_P2M_ENTRY;
+
+	if (cmpxchg(pp, p2m_missing, p) != p2m_missing)
+		free_page((unsigned long)p);
+	else
+		*mfnp = virt_to_mfn(p);
+}
+
+void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+	unsigned topidx, idx;
+
+	if (unlikely(xen_feature(XENFEAT_auto_translated_physmap))) {
+		BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
+		return;
+	}
+
+	if (unlikely(pfn >= MAX_DOMAIN_PAGES)) {
+		BUG_ON(mfn != INVALID_P2M_ENTRY);
+		return;
+	}
+
+	topidx = p2m_top_index(pfn);
+	if (p2m_top[topidx] == p2m_missing) {
+		/* no need to allocate a page to store an invalid entry */
+		if (mfn == INVALID_P2M_ENTRY)
+			return;
+		alloc_p2m(&p2m_top[topidx], &p2m_top_mfn[topidx]);
+	}
+
+	idx = p2m_index(pfn);
+	p2m_top[topidx][idx] = mfn;
+}
+
+xmaddr_t arbitrary_virt_to_machine(void *vaddr)
+{
+	unsigned long address = (unsigned long)vaddr;
 	unsigned int level;
 	pte_t *pte = lookup_address(address, &level);
 	unsigned offset = address & ~PAGE_MASK;
 
 	BUG_ON(pte == NULL);
 
-	return XMADDR((pte_mfn(*pte) << PAGE_SHIFT) + offset);
+	return XMADDR(((phys_addr_t)pte_mfn(*pte) << PAGE_SHIFT) + offset);
 }
 
 void make_lowmem_page_readonly(void *vaddr)
@@ -98,59 +285,84 @@ void make_lowmem_page_readwrite(void *vaddr)
 }
 
 
-void xen_set_pmd(pmd_t *ptr, pmd_t val)
+static bool xen_page_pinned(void *ptr)
+{
+	struct page *page = virt_to_page(ptr);
+
+	return PagePinned(page);
+}
+
+static void xen_extend_mmu_update(const struct mmu_update *update)
 {
 	struct multicall_space mcs;
 	struct mmu_update *u;
 
-	preempt_disable();
+	mcs = xen_mc_extend_args(__HYPERVISOR_mmu_update, sizeof(*u));
+
+	if (mcs.mc != NULL) {
+		ADD_STATS(mmu_update_extended, 1);
+		ADD_STATS(mmu_update_histo[mcs.mc->args[1]], -1);
+
+		mcs.mc->args[1]++;
+
+		if (mcs.mc->args[1] < MMU_UPDATE_HISTO)
+			ADD_STATS(mmu_update_histo[mcs.mc->args[1]], 1);
+		else
+			ADD_STATS(mmu_update_histo[0], 1);
+	} else {
+		ADD_STATS(mmu_update, 1);
+		mcs = __xen_mc_entry(sizeof(*u));
+		MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
+		ADD_STATS(mmu_update_histo[1], 1);
+	}
 
-	mcs = xen_mc_entry(sizeof(*u));
 	u = mcs.args;
-	u->ptr = virt_to_machine(ptr).maddr;
-	u->val = pmd_val_ma(val);
-	MULTI_mmu_update(mcs.mc, u, 1, NULL, DOMID_SELF);
+	*u = *update;
+}
+
+void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val)
+{
+	struct mmu_update u;
+
+	preempt_disable();
+
+	xen_mc_batch();
+
+	/* ptr may be ioremapped for 64-bit pagetable setup */
+	u.ptr = arbitrary_virt_to_machine(ptr).maddr;
+	u.val = pmd_val_ma(val);
+	xen_extend_mmu_update(&u);
+
+	ADD_STATS(pmd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
 
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
 
 	preempt_enable();
 }
 
+void xen_set_pmd(pmd_t *ptr, pmd_t val)
+{
+	ADD_STATS(pmd_update, 1);
+
+	/* If page is not pinned, we can just update the entry
+	   directly */
+	if (!xen_page_pinned(ptr)) {
+		*ptr = val;
+		return;
+	}
+
+	ADD_STATS(pmd_update_pinned, 1);
+
+	xen_set_pmd_hyper(ptr, val);
+}
+
 /*
  * Associate a virtual page frame with a given physical page frame
  * and protection flags for that frame.
  */
 void set_pte_mfn(unsigned long vaddr, unsigned long mfn, pgprot_t flags)
 {
-	pgd_t *pgd;
-	pud_t *pud;
-	pmd_t *pmd;
-	pte_t *pte;
-
-	pgd = swapper_pg_dir + pgd_index(vaddr);
-	if (pgd_none(*pgd)) {
-		BUG();
-		return;
-	}
-	pud = pud_offset(pgd, vaddr);
-	if (pud_none(*pud)) {
-		BUG();
-		return;
-	}
-	pmd = pmd_offset(pud, vaddr);
-	if (pmd_none(*pmd)) {
-		BUG();
-		return;
-	}
-	pte = pte_offset_kernel(pmd, vaddr);
-	/* <mfn,flags> stored as-is, to permit clearing entries */
-	xen_set_pte(pte, mfn_pte(mfn, flags));
-
-	/*
-	 * It's enough to flush this one mapping.
-	 * (PGE mappings get flushed as well)
-	 */
-	__flush_tlb_one(vaddr);
+	set_pte_vaddr(vaddr, mfn_pte(mfn, flags));
 }
 
 void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
@@ -160,12 +372,18 @@ void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
 	if (mm == &init_mm)
 		preempt_disable();
 
+	ADD_STATS(set_pte_at, 1);
+//	ADD_STATS(set_pte_at_pinned, xen_page_pinned(ptep));
+	ADD_STATS(set_pte_at_current, mm == current->mm);
+	ADD_STATS(set_pte_at_kernel, mm == &init_mm);
+
 	if (mm == current->mm || mm == &init_mm) {
 		if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) {
 			struct multicall_space mcs;
 			mcs = xen_mc_entry(0);
 
 			MULTI_update_va_mapping(mcs.mc, addr, pteval, 0);
+			ADD_STATS(set_pte_at_batched, 1);
 			xen_mc_issue(PARAVIRT_LAZY_MMU);
 			goto out;
 		} else
@@ -179,13 +397,36 @@ out:
 		preempt_enable();
 }
 
+pte_t xen_ptep_modify_prot_start(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+	/* Just return the pte as-is.  We preserve the bits on commit */
+	return *ptep;
+}
+
+void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr,
+				 pte_t *ptep, pte_t pte)
+{
+	struct mmu_update u;
+
+	xen_mc_batch();
+
+	u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD;
+	u.val = pte_val_ma(pte);
+	xen_extend_mmu_update(&u);
+
+	ADD_STATS(prot_commit, 1);
+	ADD_STATS(prot_commit_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
+
+	xen_mc_issue(PARAVIRT_LAZY_MMU);
+}
+
 /* Assume pteval_t is equivalent to all the other *val_t types. */
 static pteval_t pte_mfn_to_pfn(pteval_t val)
 {
 	if (val & _PAGE_PRESENT) {
-		unsigned long mfn = (val & PTE_MASK) >> PAGE_SHIFT;
-		pteval_t flags = val & ~PTE_MASK;
-		val = (mfn_to_pfn(mfn) << PAGE_SHIFT) | flags;
+		unsigned long mfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT;
+		pteval_t flags = val & PTE_FLAGS_MASK;
+		val = ((pteval_t)mfn_to_pfn(mfn) << PAGE_SHIFT) | flags;
 	}
 
 	return val;
@@ -194,9 +435,9 @@ static pteval_t pte_mfn_to_pfn(pteval_t val)
 static pteval_t pte_pfn_to_mfn(pteval_t val)
 {
 	if (val & _PAGE_PRESENT) {
-		unsigned long pfn = (val & PTE_MASK) >> PAGE_SHIFT;
-		pteval_t flags = val & ~PTE_MASK;
-		val = (pfn_to_mfn(pfn) << PAGE_SHIFT) | flags;
+		unsigned long pfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT;
+		pteval_t flags = val & PTE_FLAGS_MASK;
+		val = ((pteval_t)pfn_to_mfn(pfn) << PAGE_SHIFT) | flags;
 	}
 
 	return val;
@@ -229,34 +470,61 @@ pmdval_t xen_pmd_val(pmd_t pmd)
 	return pte_mfn_to_pfn(pmd.pmd);
 }
 
-void xen_set_pud(pud_t *ptr, pud_t val)
+void xen_set_pud_hyper(pud_t *ptr, pud_t val)
 {
-	struct multicall_space mcs;
-	struct mmu_update *u;
+	struct mmu_update u;
 
 	preempt_disable();
 
-	mcs = xen_mc_entry(sizeof(*u));
-	u = mcs.args;
-	u->ptr = virt_to_machine(ptr).maddr;
-	u->val = pud_val_ma(val);
-	MULTI_mmu_update(mcs.mc, u, 1, NULL, DOMID_SELF);
+	xen_mc_batch();
+
+	/* ptr may be ioremapped for 64-bit pagetable setup */
+	u.ptr = arbitrary_virt_to_machine(ptr).maddr;
+	u.val = pud_val_ma(val);
+	xen_extend_mmu_update(&u);
+
+	ADD_STATS(pud_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
 
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
 
 	preempt_enable();
 }
 
+void xen_set_pud(pud_t *ptr, pud_t val)
+{
+	ADD_STATS(pud_update, 1);
+
+	/* If page is not pinned, we can just update the entry
+	   directly */
+	if (!xen_page_pinned(ptr)) {
+		*ptr = val;
+		return;
+	}
+
+	ADD_STATS(pud_update_pinned, 1);
+
+	xen_set_pud_hyper(ptr, val);
+}
+
 void xen_set_pte(pte_t *ptep, pte_t pte)
 {
+	ADD_STATS(pte_update, 1);
+//	ADD_STATS(pte_update_pinned, xen_page_pinned(ptep));
+	ADD_STATS(pte_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
+
+#ifdef CONFIG_X86_PAE
 	ptep->pte_high = pte.pte_high;
 	smp_wmb();
 	ptep->pte_low = pte.pte_low;
+#else
+	*ptep = pte;
+#endif
 }
 
+#ifdef CONFIG_X86_PAE
 void xen_set_pte_atomic(pte_t *ptep, pte_t pte)
 {
-	set_64bit((u64 *)ptep, pte_val_ma(pte));
+	set_64bit((u64 *)ptep, native_pte_val(pte));
 }
 
 void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
@@ -268,8 +536,9 @@ void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 
 void xen_pmd_clear(pmd_t *pmdp)
 {
-	xen_set_pmd(pmdp, __pmd(0));
+	set_pmd(pmdp, __pmd(0));
 }
+#endif	/* CONFIG_X86_PAE */
 
 pmd_t xen_make_pmd(pmdval_t pmd)
 {
@@ -277,95 +546,218 @@ pmd_t xen_make_pmd(pmdval_t pmd)
 	return native_make_pmd(pmd);
 }
 
+#if PAGETABLE_LEVELS == 4
+pudval_t xen_pud_val(pud_t pud)
+{
+	return pte_mfn_to_pfn(pud.pud);
+}
+
+pud_t xen_make_pud(pudval_t pud)
+{
+	pud = pte_pfn_to_mfn(pud);
+
+	return native_make_pud(pud);
+}
+
+pgd_t *xen_get_user_pgd(pgd_t *pgd)
+{
+	pgd_t *pgd_page = (pgd_t *)(((unsigned long)pgd) & PAGE_MASK);
+	unsigned offset = pgd - pgd_page;
+	pgd_t *user_ptr = NULL;
+
+	if (offset < pgd_index(USER_LIMIT)) {
+		struct page *page = virt_to_page(pgd_page);
+		user_ptr = (pgd_t *)page->private;
+		if (user_ptr)
+			user_ptr += offset;
+	}
+
+	return user_ptr;
+}
+
+static void __xen_set_pgd_hyper(pgd_t *ptr, pgd_t val)
+{
+	struct mmu_update u;
+
+	u.ptr = virt_to_machine(ptr).maddr;
+	u.val = pgd_val_ma(val);
+	xen_extend_mmu_update(&u);
+}
+
+/*
+ * Raw hypercall-based set_pgd, intended for in early boot before
+ * there's a page structure.  This implies:
+ *  1. The only existing pagetable is the kernel's
+ *  2. It is always pinned
+ *  3. It has no user pagetable attached to it
+ */
+void __init xen_set_pgd_hyper(pgd_t *ptr, pgd_t val)
+{
+	preempt_disable();
+
+	xen_mc_batch();
+
+	__xen_set_pgd_hyper(ptr, val);
+
+	xen_mc_issue(PARAVIRT_LAZY_MMU);
+
+	preempt_enable();
+}
+
+void xen_set_pgd(pgd_t *ptr, pgd_t val)
+{
+	pgd_t *user_ptr = xen_get_user_pgd(ptr);
+
+	ADD_STATS(pgd_update, 1);
+
+	/* If page is not pinned, we can just update the entry
+	   directly */
+	if (!xen_page_pinned(ptr)) {
+		*ptr = val;
+		if (user_ptr) {
+			WARN_ON(xen_page_pinned(user_ptr));
+			*user_ptr = val;
+		}
+		return;
+	}
+
+	ADD_STATS(pgd_update_pinned, 1);
+	ADD_STATS(pgd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
+
+	/* If it's pinned, then we can at least batch the kernel and
+	   user updates together. */
+	xen_mc_batch();
+
+	__xen_set_pgd_hyper(ptr, val);
+	if (user_ptr)
+		__xen_set_pgd_hyper(user_ptr, val);
+
+	xen_mc_issue(PARAVIRT_LAZY_MMU);
+}
+#endif	/* PAGETABLE_LEVELS == 4 */
+
 /*
-  (Yet another) pagetable walker.  This one is intended for pinning a
-  pagetable.  This means that it walks a pagetable and calls the
-  callback function on each page it finds making up the page table,
-  at every level.  It walks the entire pagetable, but it only bothers
-  pinning pte pages which are below pte_limit.  In the normal case
-  this will be TASK_SIZE, but at boot we need to pin up to
-  FIXADDR_TOP.  But the important bit is that we don't pin beyond
-  there, because then we start getting into Xen's ptes.
-*/
-static int pgd_walk(pgd_t *pgd_base, int (*func)(struct page *, enum pt_level),
-		    unsigned long limit)
-{
-	pgd_t *pgd = pgd_base;
+ * (Yet another) pagetable walker.  This one is intended for pinning a
+ * pagetable.  This means that it walks a pagetable and calls the
+ * callback function on each page it finds making up the page table,
+ * at every level.  It walks the entire pagetable, but it only bothers
+ * pinning pte pages which are below limit.  In the normal case this
+ * will be STACK_TOP_MAX, but at boot we need to pin up to
+ * FIXADDR_TOP.
+ *
+ * For 32-bit the important bit is that we don't pin beyond there,
+ * because then we start getting into Xen's ptes.
+ *
+ * For 64-bit, we must skip the Xen hole in the middle of the address
+ * space, just after the big x86-64 virtual hole.
+ */
+static int xen_pgd_walk(struct mm_struct *mm,
+			int (*func)(struct mm_struct *mm, struct page *,
+				    enum pt_level),
+			unsigned long limit)
+{
+	pgd_t *pgd = mm->pgd;
 	int flush = 0;
-	unsigned long addr = 0;
-	unsigned long pgd_next;
+	unsigned hole_low, hole_high;
+	unsigned pgdidx_limit, pudidx_limit, pmdidx_limit;
+	unsigned pgdidx, pudidx, pmdidx;
 
-	BUG_ON(limit > FIXADDR_TOP);
+	/* The limit is the last byte to be touched */
+	limit--;
+	BUG_ON(limit >= FIXADDR_TOP);
 
 	if (xen_feature(XENFEAT_auto_translated_physmap))
 		return 0;
 
-	for (; addr != FIXADDR_TOP; pgd++, addr = pgd_next) {
+	/*
+	 * 64-bit has a great big hole in the middle of the address
+	 * space, which contains the Xen mappings.  On 32-bit these
+	 * will end up making a zero-sized hole and so is a no-op.
+	 */
+	hole_low = pgd_index(USER_LIMIT);
+	hole_high = pgd_index(PAGE_OFFSET);
+
+	pgdidx_limit = pgd_index(limit);
+#if PTRS_PER_PUD > 1
+	pudidx_limit = pud_index(limit);
+#else
+	pudidx_limit = 0;
+#endif
+#if PTRS_PER_PMD > 1
+	pmdidx_limit = pmd_index(limit);
+#else
+	pmdidx_limit = 0;
+#endif
+
+	for (pgdidx = 0; pgdidx <= pgdidx_limit; pgdidx++) {
 		pud_t *pud;
-		unsigned long pud_limit, pud_next;
 
-		pgd_next = pud_limit = pgd_addr_end(addr, FIXADDR_TOP);
+		if (pgdidx >= hole_low && pgdidx < hole_high)
+			continue;
 
-		if (!pgd_val(*pgd))
+		if (!pgd_val(pgd[pgdidx]))
 			continue;
 
-		pud = pud_offset(pgd, 0);
+		pud = pud_offset(&pgd[pgdidx], 0);
 
 		if (PTRS_PER_PUD > 1) /* not folded */
-			flush |= (*func)(virt_to_page(pud), PT_PUD);
+			flush |= (*func)(mm, virt_to_page(pud), PT_PUD);
 
-		for (; addr != pud_limit; pud++, addr = pud_next) {
+		for (pudidx = 0; pudidx < PTRS_PER_PUD; pudidx++) {
 			pmd_t *pmd;
-			unsigned long pmd_limit;
-
-			pud_next = pud_addr_end(addr, pud_limit);
 
-			if (pud_next < limit)
-				pmd_limit = pud_next;
-			else
-				pmd_limit = limit;
+			if (pgdidx == pgdidx_limit &&
+			    pudidx > pudidx_limit)
+				goto out;
 
-			if (pud_none(*pud))
+			if (pud_none(pud[pudidx]))
 				continue;
 
-			pmd = pmd_offset(pud, 0);
+			pmd = pmd_offset(&pud[pudidx], 0);
 
 			if (PTRS_PER_PMD > 1) /* not folded */
-				flush |= (*func)(virt_to_page(pmd), PT_PMD);
+				flush |= (*func)(mm, virt_to_page(pmd), PT_PMD);
+
+			for (pmdidx = 0; pmdidx < PTRS_PER_PMD; pmdidx++) {
+				struct page *pte;
 
-			for (; addr != pmd_limit; pmd++) {
-				addr += (PAGE_SIZE * PTRS_PER_PTE);
-				if ((pmd_limit-1) < (addr-1)) {
-					addr = pmd_limit;
-					break;
-				}
+				if (pgdidx == pgdidx_limit &&
+				    pudidx == pudidx_limit &&
+				    pmdidx > pmdidx_limit)
+					goto out;
 
-				if (pmd_none(*pmd))
+				if (pmd_none(pmd[pmdidx]))
 					continue;
 
-				flush |= (*func)(pmd_page(*pmd), PT_PTE);
+				pte = pmd_page(pmd[pmdidx]);
+				flush |= (*func)(mm, pte, PT_PTE);
 			}
 		}
 	}
 
-	flush |= (*func)(virt_to_page(pgd_base), PT_PGD);
+out:
+	/* Do the top level last, so that the callbacks can use it as
+	   a cue to do final things like tlb flushes. */
+	flush |= (*func)(mm, virt_to_page(pgd), PT_PGD);
 
 	return flush;
 }
 
-static spinlock_t *lock_pte(struct page *page)
+/* If we're using split pte locks, then take the page's lock and
+   return a pointer to it.  Otherwise return NULL. */
+static spinlock_t *xen_pte_lock(struct page *page, struct mm_struct *mm)
 {
 	spinlock_t *ptl = NULL;
 
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+#if USE_SPLIT_PTLOCKS
 	ptl = __pte_lockptr(page);
-	spin_lock(ptl);
+	spin_lock_nest_lock(ptl, &mm->page_table_lock);
 #endif
 
 	return ptl;
 }
 
-static void do_unlock(void *v)
+static void xen_pte_unlock(void *v)
 {
 	spinlock_t *ptl = v;
 	spin_unlock(ptl);
@@ -383,7 +775,8 @@ static void xen_do_pin(unsigned level, unsigned long pfn)
 	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
 }
 
-static int pin_page(struct page *page, enum pt_level level)
+static int xen_pin_page(struct mm_struct *mm, struct page *page,
+			enum pt_level level)
 {
 	unsigned pgfl = TestSetPagePinned(page);
 	int flush;
@@ -402,21 +795,40 @@ static int pin_page(struct page *page, enum pt_level level)
 
 		flush = 0;
 
+		/*
+		 * We need to hold the pagetable lock between the time
+		 * we make the pagetable RO and when we actually pin
+		 * it.  If we don't, then other users may come in and
+		 * attempt to update the pagetable by writing it,
+		 * which will fail because the memory is RO but not
+		 * pinned, so Xen won't do the trap'n'emulate.
+		 *
+		 * If we're using split pte locks, we can't hold the
+		 * entire pagetable's worth of locks during the
+		 * traverse, because we may wrap the preempt count (8
+		 * bits).  The solution is to mark RO and pin each PTE
+		 * page while holding the lock.  This means the number
+		 * of locks we end up holding is never more than a
+		 * batch size (~32 entries, at present).
+		 *
+		 * If we're not using split pte locks, we needn't pin
+		 * the PTE pages independently, because we're
+		 * protected by the overall pagetable lock.
+		 */
 		ptl = NULL;
 		if (level == PT_PTE)
-			ptl = lock_pte(page);
+			ptl = xen_pte_lock(page, mm);
 
 		MULTI_update_va_mapping(mcs.mc, (unsigned long)pt,
 					pfn_pte(pfn, PAGE_KERNEL_RO),
 					level == PT_PGD ? UVMF_TLB_FLUSH : 0);
 
-		if (level == PT_PTE)
+		if (ptl) {
 			xen_do_pin(MMUEXT_PIN_L1_TABLE, pfn);
 
-		if (ptl) {
 			/* Queue a deferred unlock for when this batch
 			   is completed. */
-			xen_mc_callback(do_unlock, ptl);
+			xen_mc_callback(xen_pte_unlock, ptl);
 		}
 	}
 
@@ -426,25 +838,78 @@ static int pin_page(struct page *page, enum pt_level level)
 /* This is called just after a mm has been created, but it has not
    been used yet.  We need to make sure that its pagetable is all
    read-only, and can be pinned. */
-void xen_pgd_pin(pgd_t *pgd)
+static void __xen_pgd_pin(struct mm_struct *mm, pgd_t *pgd)
 {
 	xen_mc_batch();
 
-	if (pgd_walk(pgd, pin_page, TASK_SIZE)) {
+	if (xen_pgd_walk(mm, xen_pin_page, USER_LIMIT)) {
 		/* re-enable interrupts for kmap_flush_unused */
 		xen_mc_issue(0);
 		kmap_flush_unused();
 		xen_mc_batch();
 	}
 
+#ifdef CONFIG_X86_64
+	{
+		pgd_t *user_pgd = xen_get_user_pgd(pgd);
+
+		xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(pgd)));
+
+		if (user_pgd) {
+			xen_pin_page(mm, virt_to_page(user_pgd), PT_PGD);
+			xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(user_pgd)));
+		}
+	}
+#else /* CONFIG_X86_32 */
+#ifdef CONFIG_X86_PAE
+	/* Need to make sure unshared kernel PMD is pinnable */
+	xen_pin_page(mm, virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])),
+		     PT_PMD);
+#endif
 	xen_do_pin(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(pgd)));
+#endif /* CONFIG_X86_64 */
 	xen_mc_issue(0);
 }
 
-/* The init_mm pagetable is really pinned as soon as its created, but
-   that's before we have page structures to store the bits.  So do all
-   the book-keeping now. */
-static __init int mark_pinned(struct page *page, enum pt_level level)
+static void xen_pgd_pin(struct mm_struct *mm)
+{
+	__xen_pgd_pin(mm, mm->pgd);
+}
+
+/*
+ * On save, we need to pin all pagetables to make sure they get their
+ * mfns turned into pfns.  Search the list for any unpinned pgds and pin
+ * them (unpinned pgds are not currently in use, probably because the
+ * process is under construction or destruction).
+ *
+ * Expected to be called in stop_machine() ("equivalent to taking
+ * every spinlock in the system"), so the locking doesn't really
+ * matter all that much.
+ */
+void xen_mm_pin_all(void)
+{
+	unsigned long flags;
+	struct page *page;
+
+	spin_lock_irqsave(&pgd_lock, flags);
+
+	list_for_each_entry(page, &pgd_list, lru) {
+		if (!PagePinned(page)) {
+			__xen_pgd_pin(&init_mm, (pgd_t *)page_address(page));
+			SetPageSavePinned(page);
+		}
+	}
+
+	spin_unlock_irqrestore(&pgd_lock, flags);
+}
+
+/*
+ * The init_mm pagetable is really pinned as soon as its created, but
+ * that's before we have page structures to store the bits.  So do all
+ * the book-keeping now.
+ */
+static __init int xen_mark_pinned(struct mm_struct *mm, struct page *page,
+				  enum pt_level level)
 {
 	SetPagePinned(page);
 	return 0;
@@ -452,10 +917,11 @@ static __init int mark_pinned(struct page *page, enum pt_level level)
 
 void __init xen_mark_init_mm_pinned(void)
 {
-	pgd_walk(init_mm.pgd, mark_pinned, FIXADDR_TOP);
+	xen_pgd_walk(&init_mm, xen_mark_pinned, FIXADDR_TOP);
 }
 
-static int unpin_page(struct page *page, enum pt_level level)
+static int xen_unpin_page(struct mm_struct *mm, struct page *page,
+			  enum pt_level level)
 {
 	unsigned pgfl = TestClearPagePinned(page);
 
@@ -465,10 +931,18 @@ static int unpin_page(struct page *page, enum pt_level level)
 		spinlock_t *ptl = NULL;
 		struct multicall_space mcs;
 
+		/*
+		 * Do the converse to pin_page.  If we're using split
+		 * pte locks, we must be holding the lock for while
+		 * the pte page is unpinned but still RO to prevent
+		 * concurrent updates from seeing it in this
+		 * partially-pinned state.
+		 */
 		if (level == PT_PTE) {
-			ptl = lock_pte(page);
+			ptl = xen_pte_lock(page, mm);
 
-			xen_do_pin(MMUEXT_UNPIN_TABLE, pfn);
+			if (ptl)
+				xen_do_pin(MMUEXT_UNPIN_TABLE, pfn);
 		}
 
 		mcs = __xen_mc_entry(0);
@@ -479,7 +953,7 @@ static int unpin_page(struct page *page, enum pt_level level)
 
 		if (ptl) {
 			/* unlock when batch completed */
-			xen_mc_callback(do_unlock, ptl);
+			xen_mc_callback(xen_pte_unlock, ptl);
 		}
 	}
 
@@ -487,28 +961,72 @@ static int unpin_page(struct page *page, enum pt_level level)
 }
 
 /* Release a pagetables pages back as normal RW */
-static void xen_pgd_unpin(pgd_t *pgd)
+static void __xen_pgd_unpin(struct mm_struct *mm, pgd_t *pgd)
 {
 	xen_mc_batch();
 
 	xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
 
-	pgd_walk(pgd, unpin_page, TASK_SIZE);
+#ifdef CONFIG_X86_64
+	{
+		pgd_t *user_pgd = xen_get_user_pgd(pgd);
+
+		if (user_pgd) {
+			xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(user_pgd)));
+			xen_unpin_page(mm, virt_to_page(user_pgd), PT_PGD);
+		}
+	}
+#endif
+
+#ifdef CONFIG_X86_PAE
+	/* Need to make sure unshared kernel PMD is unpinned */
+	xen_unpin_page(mm, virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])),
+		       PT_PMD);
+#endif
+
+	xen_pgd_walk(mm, xen_unpin_page, USER_LIMIT);
 
 	xen_mc_issue(0);
 }
 
+static void xen_pgd_unpin(struct mm_struct *mm)
+{
+	__xen_pgd_unpin(mm, mm->pgd);
+}
+
+/*
+ * On resume, undo any pinning done at save, so that the rest of the
+ * kernel doesn't see any unexpected pinned pagetables.
+ */
+void xen_mm_unpin_all(void)
+{
+	unsigned long flags;
+	struct page *page;
+
+	spin_lock_irqsave(&pgd_lock, flags);
+
+	list_for_each_entry(page, &pgd_list, lru) {
+		if (PageSavePinned(page)) {
+			BUG_ON(!PagePinned(page));
+			__xen_pgd_unpin(&init_mm, (pgd_t *)page_address(page));
+			ClearPageSavePinned(page);
+		}
+	}
+
+	spin_unlock_irqrestore(&pgd_lock, flags);
+}
+
 void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next)
 {
 	spin_lock(&next->page_table_lock);
-	xen_pgd_pin(next->pgd);
+	xen_pgd_pin(next);
 	spin_unlock(&next->page_table_lock);
 }
 
 void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
 {
 	spin_lock(&mm->page_table_lock);
-	xen_pgd_pin(mm->pgd);
+	xen_pgd_pin(mm);
 	spin_unlock(&mm->page_table_lock);
 }
 
@@ -519,8 +1037,15 @@ void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
 static void drop_other_mm_ref(void *info)
 {
 	struct mm_struct *mm = info;
+	struct mm_struct *active_mm;
+
+#ifdef CONFIG_X86_64
+	active_mm = read_pda(active_mm);
+#else
+	active_mm = __get_cpu_var(cpu_tlbstate).active_mm;
+#endif
 
-	if (__get_cpu_var(cpu_tlbstate).active_mm == mm)
+	if (active_mm == mm)
 		leave_mm(smp_processor_id());
 
 	/* If this cpu still has a stale cr3 reference, then make sure
@@ -531,7 +1056,7 @@ static void drop_other_mm_ref(void *info)
 	}
 }
 
-static void drop_mm_ref(struct mm_struct *mm)
+static void xen_drop_mm_ref(struct mm_struct *mm)
 {
 	cpumask_t mask;
 	unsigned cpu;
@@ -558,10 +1083,10 @@ static void drop_mm_ref(struct mm_struct *mm)
 	}
 
 	if (!cpus_empty(mask))
-		xen_smp_call_function_mask(mask, drop_other_mm_ref, mm, 1);
+		smp_call_function_mask(mask, drop_other_mm_ref, mm, 1);
 }
 #else
-static void drop_mm_ref(struct mm_struct *mm)
+static void xen_drop_mm_ref(struct mm_struct *mm)
 {
 	if (current->active_mm == mm)
 		load_cr3(swapper_pg_dir);
@@ -585,14 +1110,77 @@ static void drop_mm_ref(struct mm_struct *mm)
 void xen_exit_mmap(struct mm_struct *mm)
 {
 	get_cpu();		/* make sure we don't move around */
-	drop_mm_ref(mm);
+	xen_drop_mm_ref(mm);
 	put_cpu();
 
 	spin_lock(&mm->page_table_lock);
 
 	/* pgd may not be pinned in the error exit path of execve */
-	if (PagePinned(virt_to_page(mm->pgd)))
-		xen_pgd_unpin(mm->pgd);
+	if (xen_page_pinned(mm->pgd))
+		xen_pgd_unpin(mm);
 
 	spin_unlock(&mm->page_table_lock);
 }
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct dentry *d_mmu_debug;
+
+static int __init xen_mmu_debugfs(void)
+{
+	struct dentry *d_xen = xen_init_debugfs();
+
+	if (d_xen == NULL)
+		return -ENOMEM;
+
+	d_mmu_debug = debugfs_create_dir("mmu", d_xen);
+
+	debugfs_create_u8("zero_stats", 0644, d_mmu_debug, &zero_stats);
+
+	debugfs_create_u32("pgd_update", 0444, d_mmu_debug, &mmu_stats.pgd_update);
+	debugfs_create_u32("pgd_update_pinned", 0444, d_mmu_debug,
+			   &mmu_stats.pgd_update_pinned);
+	debugfs_create_u32("pgd_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pgd_update_pinned);
+
+	debugfs_create_u32("pud_update", 0444, d_mmu_debug, &mmu_stats.pud_update);
+	debugfs_create_u32("pud_update_pinned", 0444, d_mmu_debug,
+			   &mmu_stats.pud_update_pinned);
+	debugfs_create_u32("pud_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pud_update_pinned);
+
+	debugfs_create_u32("pmd_update", 0444, d_mmu_debug, &mmu_stats.pmd_update);
+	debugfs_create_u32("pmd_update_pinned", 0444, d_mmu_debug,
+			   &mmu_stats.pmd_update_pinned);
+	debugfs_create_u32("pmd_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pmd_update_pinned);
+
+	debugfs_create_u32("pte_update", 0444, d_mmu_debug, &mmu_stats.pte_update);
+//	debugfs_create_u32("pte_update_pinned", 0444, d_mmu_debug,
+//			   &mmu_stats.pte_update_pinned);
+	debugfs_create_u32("pte_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pte_update_pinned);
+
+	debugfs_create_u32("mmu_update", 0444, d_mmu_debug, &mmu_stats.mmu_update);
+	debugfs_create_u32("mmu_update_extended", 0444, d_mmu_debug,
+			   &mmu_stats.mmu_update_extended);
+	xen_debugfs_create_u32_array("mmu_update_histo", 0444, d_mmu_debug,
+				     mmu_stats.mmu_update_histo, 20);
+
+	debugfs_create_u32("set_pte_at", 0444, d_mmu_debug, &mmu_stats.set_pte_at);
+	debugfs_create_u32("set_pte_at_batched", 0444, d_mmu_debug,
+			   &mmu_stats.set_pte_at_batched);
+	debugfs_create_u32("set_pte_at_current", 0444, d_mmu_debug,
+			   &mmu_stats.set_pte_at_current);
+	debugfs_create_u32("set_pte_at_kernel", 0444, d_mmu_debug,
+			   &mmu_stats.set_pte_at_kernel);
+
+	debugfs_create_u32("prot_commit", 0444, d_mmu_debug, &mmu_stats.prot_commit);
+	debugfs_create_u32("prot_commit_batched", 0444, d_mmu_debug,
+			   &mmu_stats.prot_commit_batched);
+
+	return 0;
+}
+fs_initcall(xen_mmu_debugfs);
+
+#endif	/* CONFIG_XEN_DEBUG_FS */
diff --git a/arch/x86/xen/mmu.h b/arch/x86/xen/mmu.h
index 5fe961caffd..98d71659da5 100644
--- a/arch/x86/xen/mmu.h
+++ b/arch/x86/xen/mmu.h
@@ -10,33 +10,14 @@ enum pt_level {
 	PT_PTE
 };
 
-/*
- * Page-directory addresses above 4GB do not fit into architectural %cr3.
- * When accessing %cr3, or equivalent field in vcpu_guest_context, guests
- * must use the following accessor macros to pack/unpack valid MFNs.
- *
- * Note that Xen is using the fact that the pagetable base is always
- * page-aligned, and putting the 12 MSB of the address into the 12 LSB
- * of cr3.
- */
-#define xen_pfn_to_cr3(pfn) (((unsigned)(pfn) << 12) | ((unsigned)(pfn) >> 20))
-#define xen_cr3_to_pfn(cr3) (((unsigned)(cr3) >> 12) | ((unsigned)(cr3) << 20))
-
 
 void set_pte_mfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags);
 
-void xen_set_pte(pte_t *ptep, pte_t pteval);
-void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
-		    pte_t *ptep, pte_t pteval);
-void xen_set_pmd(pmd_t *pmdp, pmd_t pmdval);
 
 void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next);
 void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm);
 void xen_exit_mmap(struct mm_struct *mm);
 
-void xen_pgd_pin(pgd_t *pgd);
-//void xen_pgd_unpin(pgd_t *pgd);
-
 pteval_t xen_pte_val(pte_t);
 pmdval_t xen_pmd_val(pmd_t);
 pgdval_t xen_pgd_val(pgd_t);
@@ -45,11 +26,32 @@ pte_t xen_make_pte(pteval_t);
 pmd_t xen_make_pmd(pmdval_t);
 pgd_t xen_make_pgd(pgdval_t);
 
+void xen_set_pte(pte_t *ptep, pte_t pteval);
 void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
 		    pte_t *ptep, pte_t pteval);
+
+#ifdef CONFIG_X86_PAE
 void xen_set_pte_atomic(pte_t *ptep, pte_t pte);
-void xen_set_pud(pud_t *ptr, pud_t val);
 void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
 void xen_pmd_clear(pmd_t *pmdp);
+#endif	/* CONFIG_X86_PAE */
+
+void xen_set_pmd(pmd_t *pmdp, pmd_t pmdval);
+void xen_set_pud(pud_t *ptr, pud_t val);
+void xen_set_pmd_hyper(pmd_t *pmdp, pmd_t pmdval);
+void xen_set_pud_hyper(pud_t *ptr, pud_t val);
+
+#if PAGETABLE_LEVELS == 4
+pudval_t xen_pud_val(pud_t pud);
+pud_t xen_make_pud(pudval_t pudval);
+void xen_set_pgd(pgd_t *pgdp, pgd_t pgd);
+void xen_set_pgd_hyper(pgd_t *pgdp, pgd_t pgd);
+#endif
+
+pgd_t *xen_get_user_pgd(pgd_t *pgd);
+
+pte_t xen_ptep_modify_prot_start(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
+void  xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr,
+				  pte_t *ptep, pte_t pte);
 
 #endif	/* _XEN_MMU_H */
diff --git a/arch/x86/xen/multicalls.c b/arch/x86/xen/multicalls.c
index 5791eb2e375..8ea8a0d0b0d 100644
--- a/arch/x86/xen/multicalls.c
+++ b/arch/x86/xen/multicalls.c
@@ -21,22 +21,26 @@
  */
 #include <linux/percpu.h>
 #include <linux/hardirq.h>
+#include <linux/debugfs.h>
 
 #include <asm/xen/hypercall.h>
 
 #include "multicalls.h"
+#include "debugfs.h"
+
+#define MC_BATCH	32
 
 #define MC_DEBUG	1
 
-#define MC_BATCH	32
-#define MC_ARGS		(MC_BATCH * 16 / sizeof(u64))
+#define MC_ARGS		(MC_BATCH * 16)
+
 
 struct mc_buffer {
 	struct multicall_entry entries[MC_BATCH];
 #if MC_DEBUG
 	struct multicall_entry debug[MC_BATCH];
 #endif
-	u64 args[MC_ARGS];
+	unsigned char args[MC_ARGS];
 	struct callback {
 		void (*fn)(void *);
 		void *data;
@@ -47,6 +51,76 @@ struct mc_buffer {
 static DEFINE_PER_CPU(struct mc_buffer, mc_buffer);
 DEFINE_PER_CPU(unsigned long, xen_mc_irq_flags);
 
+/* flush reasons 0- slots, 1- args, 2- callbacks */
+enum flush_reasons
+{
+	FL_SLOTS,
+	FL_ARGS,
+	FL_CALLBACKS,
+
+	FL_N_REASONS
+};
+
+#ifdef CONFIG_XEN_DEBUG_FS
+#define NHYPERCALLS	40		/* not really */
+
+static struct {
+	unsigned histo[MC_BATCH+1];
+
+	unsigned issued;
+	unsigned arg_total;
+	unsigned hypercalls;
+	unsigned histo_hypercalls[NHYPERCALLS];
+
+	unsigned flush[FL_N_REASONS];
+} mc_stats;
+
+static u8 zero_stats;
+
+static inline void check_zero(void)
+{
+	if (unlikely(zero_stats)) {
+		memset(&mc_stats, 0, sizeof(mc_stats));
+		zero_stats = 0;
+	}
+}
+
+static void mc_add_stats(const struct mc_buffer *mc)
+{
+	int i;
+
+	check_zero();
+
+	mc_stats.issued++;
+	mc_stats.hypercalls += mc->mcidx;
+	mc_stats.arg_total += mc->argidx;
+
+	mc_stats.histo[mc->mcidx]++;
+	for(i = 0; i < mc->mcidx; i++) {
+		unsigned op = mc->entries[i].op;
+		if (op < NHYPERCALLS)
+			mc_stats.histo_hypercalls[op]++;
+	}
+}
+
+static void mc_stats_flush(enum flush_reasons idx)
+{
+	check_zero();
+
+	mc_stats.flush[idx]++;
+}
+
+#else  /* !CONFIG_XEN_DEBUG_FS */
+
+static inline void mc_add_stats(const struct mc_buffer *mc)
+{
+}
+
+static inline void mc_stats_flush(enum flush_reasons idx)
+{
+}
+#endif	/* CONFIG_XEN_DEBUG_FS */
+
 void xen_mc_flush(void)
 {
 	struct mc_buffer *b = &__get_cpu_var(mc_buffer);
@@ -60,6 +134,8 @@ void xen_mc_flush(void)
 	   something in the middle */
 	local_irq_save(flags);
 
+	mc_add_stats(b);
+
 	if (b->mcidx) {
 #if MC_DEBUG
 		memcpy(b->debug, b->entries,
@@ -76,6 +152,7 @@ void xen_mc_flush(void)
 		if (ret) {
 			printk(KERN_ERR "%d multicall(s) failed: cpu %d\n",
 			       ret, smp_processor_id());
+			dump_stack();
 			for (i = 0; i < b->mcidx; i++) {
 				printk("  call %2d/%d: op=%lu arg=[%lx] result=%ld\n",
 				       i+1, b->mcidx,
@@ -107,20 +184,49 @@ struct multicall_space __xen_mc_entry(size_t args)
 {
 	struct mc_buffer *b = &__get_cpu_var(mc_buffer);
 	struct multicall_space ret;
-	unsigned argspace = (args + sizeof(u64) - 1) / sizeof(u64);
+	unsigned argidx = roundup(b->argidx, sizeof(u64));
 
 	BUG_ON(preemptible());
-	BUG_ON(argspace > MC_ARGS);
+	BUG_ON(b->argidx > MC_ARGS);
 
 	if (b->mcidx == MC_BATCH ||
-	    (b->argidx + argspace) > MC_ARGS)
+	    (argidx + args) > MC_ARGS) {
+		mc_stats_flush(b->mcidx == MC_BATCH ? FL_SLOTS : FL_ARGS);
 		xen_mc_flush();
+		argidx = roundup(b->argidx, sizeof(u64));
+	}
 
 	ret.mc = &b->entries[b->mcidx];
 	b->mcidx++;
+	ret.args = &b->args[argidx];
+	b->argidx = argidx + args;
+
+	BUG_ON(b->argidx > MC_ARGS);
+	return ret;
+}
+
+struct multicall_space xen_mc_extend_args(unsigned long op, size_t size)
+{
+	struct mc_buffer *b = &__get_cpu_var(mc_buffer);
+	struct multicall_space ret = { NULL, NULL };
+
+	BUG_ON(preemptible());
+	BUG_ON(b->argidx > MC_ARGS);
+
+	if (b->mcidx == 0)
+		return ret;
+
+	if (b->entries[b->mcidx - 1].op != op)
+		return ret;
+
+	if ((b->argidx + size) > MC_ARGS)
+		return ret;
+
+	ret.mc = &b->entries[b->mcidx - 1];
 	ret.args = &b->args[b->argidx];
-	b->argidx += argspace;
+	b->argidx += size;
 
+	BUG_ON(b->argidx > MC_ARGS);
 	return ret;
 }
 
@@ -129,10 +235,44 @@ void xen_mc_callback(void (*fn)(void *), void *data)
 	struct mc_buffer *b = &__get_cpu_var(mc_buffer);
 	struct callback *cb;
 
-	if (b->cbidx == MC_BATCH)
+	if (b->cbidx == MC_BATCH) {
+		mc_stats_flush(FL_CALLBACKS);
 		xen_mc_flush();
+	}
 
 	cb = &b->callbacks[b->cbidx++];
 	cb->fn = fn;
 	cb->data = data;
 }
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct dentry *d_mc_debug;
+
+static int __init xen_mc_debugfs(void)
+{
+	struct dentry *d_xen = xen_init_debugfs();
+
+	if (d_xen == NULL)
+		return -ENOMEM;
+
+	d_mc_debug = debugfs_create_dir("multicalls", d_xen);
+
+	debugfs_create_u8("zero_stats", 0644, d_mc_debug, &zero_stats);
+
+	debugfs_create_u32("batches", 0444, d_mc_debug, &mc_stats.issued);
+	debugfs_create_u32("hypercalls", 0444, d_mc_debug, &mc_stats.hypercalls);
+	debugfs_create_u32("arg_total", 0444, d_mc_debug, &mc_stats.arg_total);
+
+	xen_debugfs_create_u32_array("batch_histo", 0444, d_mc_debug,
+				     mc_stats.histo, MC_BATCH);
+	xen_debugfs_create_u32_array("hypercall_histo", 0444, d_mc_debug,
+				     mc_stats.histo_hypercalls, NHYPERCALLS);
+	xen_debugfs_create_u32_array("flush_reasons", 0444, d_mc_debug,
+				     mc_stats.flush, FL_N_REASONS);
+
+	return 0;
+}
+fs_initcall(xen_mc_debugfs);
+
+#endif	/* CONFIG_XEN_DEBUG_FS */
diff --git a/arch/x86/xen/multicalls.h b/arch/x86/xen/multicalls.h
index 8bae996d99a..85893824161 100644
--- a/arch/x86/xen/multicalls.h
+++ b/arch/x86/xen/multicalls.h
@@ -45,4 +45,16 @@ static inline void xen_mc_issue(unsigned mode)
 /* Set up a callback to be called when the current batch is flushed */
 void xen_mc_callback(void (*fn)(void *), void *data);
 
+/*
+ * Try to extend the arguments of the previous multicall command.  The
+ * previous command's op must match.  If it does, then it attempts to
+ * extend the argument space allocated to the multicall entry by
+ * arg_size bytes.
+ *
+ * The returned multicall_space will return with mc pointing to the
+ * command on success, or NULL on failure, and args pointing to the
+ * newly allocated space.
+ */
+struct multicall_space xen_mc_extend_args(unsigned long op, size_t arg_size);
+
 #endif /* _XEN_MULTICALLS_H */
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index 82517e4a752..d6790108388 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -13,9 +13,11 @@
 #include <asm/vdso.h>
 #include <asm/e820.h>
 #include <asm/setup.h>
+#include <asm/acpi.h>
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/hypercall.h>
 
+#include <xen/page.h>
 #include <xen/interface/callback.h>
 #include <xen/interface/physdev.h>
 #include <xen/features.h>
@@ -27,8 +29,6 @@
 extern const char xen_hypervisor_callback[];
 extern const char xen_failsafe_callback[];
 
-unsigned long *phys_to_machine_mapping;
-EXPORT_SYMBOL(phys_to_machine_mapping);
 
 /**
  * machine_specific_memory_setup - Hook for machine specific memory setup.
@@ -38,9 +38,31 @@ char * __init xen_memory_setup(void)
 {
 	unsigned long max_pfn = xen_start_info->nr_pages;
 
+	max_pfn = min(MAX_DOMAIN_PAGES, max_pfn);
+
 	e820.nr_map = 0;
-	add_memory_region(0, LOWMEMSIZE(), E820_RAM);
-	add_memory_region(HIGH_MEMORY, PFN_PHYS(max_pfn)-HIGH_MEMORY, E820_RAM);
+
+	e820_add_region(0, PFN_PHYS((u64)max_pfn), E820_RAM);
+
+	/*
+	 * Even though this is normal, usable memory under Xen, reserve
+	 * ISA memory anyway because too many things think they can poke
+	 * about in there.
+	 */
+	e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS,
+			E820_RESERVED);
+
+	/*
+	 * Reserve Xen bits:
+	 *  - mfn_list
+	 *  - xen_start_info
+	 * See comment above "struct start_info" in <xen/interface/xen.h>
+	 */
+	e820_add_region(__pa(xen_start_info->mfn_list),
+			xen_start_info->pt_base - xen_start_info->mfn_list,
+			E820_RESERVED);
+
+	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
 
 	return "Xen";
 }
@@ -61,30 +83,72 @@ static void xen_idle(void)
 
 /*
  * Set the bit indicating "nosegneg" library variants should be used.
+ * We only need to bother in pure 32-bit mode; compat 32-bit processes
+ * can have un-truncated segments, so wrapping around is allowed.
  */
 static void __init fiddle_vdso(void)
 {
-	extern const char vdso32_default_start;
-	u32 *mask = VDSO32_SYMBOL(&vdso32_default_start, NOTE_MASK);
+#ifdef CONFIG_X86_32
+	u32 *mask;
+	mask = VDSO32_SYMBOL(&vdso32_int80_start, NOTE_MASK);
+	*mask |= 1 << VDSO_NOTE_NONEGSEG_BIT;
+	mask = VDSO32_SYMBOL(&vdso32_sysenter_start, NOTE_MASK);
 	*mask |= 1 << VDSO_NOTE_NONEGSEG_BIT;
+#endif
 }
 
-void xen_enable_sysenter(void)
+static __cpuinit int register_callback(unsigned type, const void *func)
 {
-	int cpu = smp_processor_id();
-	extern void xen_sysenter_target(void);
-	/* Mask events on entry, even though they get enabled immediately */
-	static struct callback_register sysenter = {
-		.type = CALLBACKTYPE_sysenter,
-		.address = { __KERNEL_CS, (unsigned long)xen_sysenter_target },
+	struct callback_register callback = {
+		.type = type,
+		.address = XEN_CALLBACK(__KERNEL_CS, func),
 		.flags = CALLBACKF_mask_events,
 	};
 
-	if (!boot_cpu_has(X86_FEATURE_SEP) ||
-	    HYPERVISOR_callback_op(CALLBACKOP_register, &sysenter) != 0) {
-		clear_cpu_cap(&cpu_data(cpu), X86_FEATURE_SEP);
-		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_SEP);
+	return HYPERVISOR_callback_op(CALLBACKOP_register, &callback);
+}
+
+void __cpuinit xen_enable_sysenter(void)
+{
+	extern void xen_sysenter_target(void);
+	int ret;
+	unsigned sysenter_feature;
+
+#ifdef CONFIG_X86_32
+	sysenter_feature = X86_FEATURE_SEP;
+#else
+	sysenter_feature = X86_FEATURE_SYSENTER32;
+#endif
+
+	if (!boot_cpu_has(sysenter_feature))
+		return;
+
+	ret = register_callback(CALLBACKTYPE_sysenter, xen_sysenter_target);
+	if(ret != 0)
+		setup_clear_cpu_cap(sysenter_feature);
+}
+
+void __cpuinit xen_enable_syscall(void)
+{
+#ifdef CONFIG_X86_64
+	int ret;
+	extern void xen_syscall_target(void);
+	extern void xen_syscall32_target(void);
+
+	ret = register_callback(CALLBACKTYPE_syscall, xen_syscall_target);
+	if (ret != 0) {
+		printk(KERN_ERR "Failed to set syscall callback: %d\n", ret);
+		/* Pretty fatal; 64-bit userspace has no other
+		   mechanism for syscalls. */
+	}
+
+	if (boot_cpu_has(X86_FEATURE_SYSCALL32)) {
+		ret = register_callback(CALLBACKTYPE_syscall32,
+					xen_syscall32_target);
+		if (ret != 0)
+			setup_clear_cpu_cap(X86_FEATURE_SYSCALL32);
 	}
+#endif /* CONFIG_X86_64 */
 }
 
 void __init xen_arch_setup(void)
@@ -98,10 +162,12 @@ void __init xen_arch_setup(void)
 	if (!xen_feature(XENFEAT_auto_translated_physmap))
 		HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_pae_extended_cr3);
 
-	HYPERVISOR_set_callbacks(__KERNEL_CS, (unsigned long)xen_hypervisor_callback,
-				 __KERNEL_CS, (unsigned long)xen_failsafe_callback);
+	if (register_callback(CALLBACKTYPE_event, xen_hypervisor_callback) ||
+	    register_callback(CALLBACKTYPE_failsafe, xen_failsafe_callback))
+		BUG();
 
 	xen_enable_sysenter();
+	xen_enable_syscall();
 
 	set_iopl.iopl = 1;
 	rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
@@ -121,11 +187,6 @@ void __init xen_arch_setup(void)
 
 	pm_idle = xen_idle;
 
-#ifdef CONFIG_SMP
-	/* fill cpus_possible with all available cpus */
-	xen_fill_possible_map();
-#endif
-
 	paravirt_disable_iospace();
 
 	fiddle_vdso();
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 94e69000f98..d77da613b1d 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -11,8 +11,6 @@
  * useful topology information for the kernel to make use of.  As a
  * result, all CPUs are treated as if they're single-core and
  * single-threaded.
- *
- * This does not handle HOTPLUG_CPU yet.
  */
 #include <linux/sched.h>
 #include <linux/err.h>
@@ -35,28 +33,15 @@
 #include "xen-ops.h"
 #include "mmu.h"
 
-static cpumask_t xen_cpu_initialized_map;
-static DEFINE_PER_CPU(int, resched_irq) = -1;
-static DEFINE_PER_CPU(int, callfunc_irq) = -1;
-static DEFINE_PER_CPU(int, debug_irq) = -1;
+cpumask_t xen_cpu_initialized_map;
 
-/*
- * Structure and data for smp_call_function(). This is designed to minimise
- * static memory requirements. It also looks cleaner.
- */
-static DEFINE_SPINLOCK(call_lock);
-
-struct call_data_struct {
-	void (*func) (void *info);
-	void *info;
-	atomic_t started;
-	atomic_t finished;
-	int wait;
-};
+static DEFINE_PER_CPU(int, resched_irq);
+static DEFINE_PER_CPU(int, callfunc_irq);
+static DEFINE_PER_CPU(int, callfuncsingle_irq);
+static DEFINE_PER_CPU(int, debug_irq) = -1;
 
 static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id);
-
-static struct call_data_struct *call_data;
+static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id);
 
 /*
  * Reschedule call back. Nothing to do,
@@ -65,25 +50,46 @@ static struct call_data_struct *call_data;
  */
 static irqreturn_t xen_reschedule_interrupt(int irq, void *dev_id)
 {
+#ifdef CONFIG_X86_32
+	__get_cpu_var(irq_stat).irq_resched_count++;
+#else
+	add_pda(irq_resched_count, 1);
+#endif
+
 	return IRQ_HANDLED;
 }
 
-static __cpuinit void cpu_bringup_and_idle(void)
+static __cpuinit void cpu_bringup(void)
 {
 	int cpu = smp_processor_id();
 
 	cpu_init();
+	touch_softlockup_watchdog();
+	preempt_disable();
+
 	xen_enable_sysenter();
+	xen_enable_syscall();
 
-	preempt_disable();
-	per_cpu(cpu_state, cpu) = CPU_ONLINE;
+	cpu = smp_processor_id();
+	smp_store_cpu_info(cpu);
+	cpu_data(cpu).x86_max_cores = 1;
+	set_cpu_sibling_map(cpu);
 
 	xen_setup_cpu_clockevents();
 
+	cpu_set(cpu, cpu_online_map);
+	x86_write_percpu(cpu_state, CPU_ONLINE);
+	wmb();
+
 	/* We can take interrupts now: we're officially "up". */
 	local_irq_enable();
 
 	wmb();			/* make sure everything is out */
+}
+
+static __cpuinit void cpu_bringup_and_idle(void)
+{
+	cpu_bringup();
 	cpu_idle();
 }
 
@@ -122,6 +128,17 @@ static int xen_smp_intr_init(unsigned int cpu)
 		goto fail;
 	per_cpu(debug_irq, cpu) = rc;
 
+	callfunc_name = kasprintf(GFP_KERNEL, "callfuncsingle%d", cpu);
+	rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_SINGLE_VECTOR,
+				    cpu,
+				    xen_call_function_single_interrupt,
+				    IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				    callfunc_name,
+				    NULL);
+	if (rc < 0)
+		goto fail;
+	per_cpu(callfuncsingle_irq, cpu) = rc;
+
 	return 0;
 
  fail:
@@ -131,59 +148,45 @@ static int xen_smp_intr_init(unsigned int cpu)
 		unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
 	if (per_cpu(debug_irq, cpu) >= 0)
 		unbind_from_irqhandler(per_cpu(debug_irq, cpu), NULL);
+	if (per_cpu(callfuncsingle_irq, cpu) >= 0)
+		unbind_from_irqhandler(per_cpu(callfuncsingle_irq, cpu), NULL);
+
 	return rc;
 }
 
-void __init xen_fill_possible_map(void)
+static void __init xen_fill_possible_map(void)
 {
 	int i, rc;
 
 	for (i = 0; i < NR_CPUS; i++) {
 		rc = HYPERVISOR_vcpu_op(VCPUOP_is_up, i, NULL);
-		if (rc >= 0)
+		if (rc >= 0) {
+			num_processors++;
 			cpu_set(i, cpu_possible_map);
+		}
 	}
 }
 
-void __init xen_smp_prepare_boot_cpu(void)
+static void __init xen_smp_prepare_boot_cpu(void)
 {
-	int cpu;
-
 	BUG_ON(smp_processor_id() != 0);
 	native_smp_prepare_boot_cpu();
 
 	/* We've switched to the "real" per-cpu gdt, so make sure the
 	   old memory can be recycled */
-	make_lowmem_page_readwrite(&per_cpu__gdt_page);
-
-	for_each_possible_cpu(cpu) {
-		cpus_clear(per_cpu(cpu_sibling_map, cpu));
-		/*
-		 * cpu_core_map lives in a per cpu area that is cleared
-		 * when the per cpu array is allocated.
-		 *
-		 * cpus_clear(per_cpu(cpu_core_map, cpu));
-		 */
-	}
+	make_lowmem_page_readwrite(&per_cpu_var(gdt_page));
 
 	xen_setup_vcpu_info_placement();
 }
 
-void __init xen_smp_prepare_cpus(unsigned int max_cpus)
+static void __init xen_smp_prepare_cpus(unsigned int max_cpus)
 {
 	unsigned cpu;
 
-	for_each_possible_cpu(cpu) {
-		cpus_clear(per_cpu(cpu_sibling_map, cpu));
-		/*
-		 * cpu_core_ map will be zeroed when the per
-		 * cpu area is allocated.
-		 *
-		 * cpus_clear(per_cpu(cpu_core_map, cpu));
-		 */
-	}
+	xen_init_lock_cpu(0);
 
 	smp_store_cpu_info(0);
+	cpu_data(0).x86_max_cores = 1;
 	set_cpu_sibling_map(0);
 
 	if (xen_smp_intr_init(0))
@@ -210,15 +213,13 @@ void __init xen_smp_prepare_cpus(unsigned int max_cpus)
 
 		cpu_set(cpu, cpu_present_map);
 	}
-
-	//init_xenbus_allowed_cpumask();
 }
 
 static __cpuinit int
 cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 {
 	struct vcpu_guest_context *ctxt;
-	struct gdt_page *gdt = &per_cpu(gdt_page, cpu);
+	struct desc_struct *gdt;
 
 	if (cpu_test_and_set(cpu, xen_cpu_initialized_map))
 		return 0;
@@ -227,12 +228,15 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 	if (ctxt == NULL)
 		return -ENOMEM;
 
+	gdt = get_cpu_gdt_table(cpu);
+
 	ctxt->flags = VGCF_IN_KERNEL;
 	ctxt->user_regs.ds = __USER_DS;
 	ctxt->user_regs.es = __USER_DS;
-	ctxt->user_regs.fs = __KERNEL_PERCPU;
-	ctxt->user_regs.gs = 0;
 	ctxt->user_regs.ss = __KERNEL_DS;
+#ifdef CONFIG_X86_32
+	ctxt->user_regs.fs = __KERNEL_PERCPU;
+#endif
 	ctxt->user_regs.eip = (unsigned long)cpu_bringup_and_idle;
 	ctxt->user_regs.eflags = 0x1000; /* IOPL_RING1 */
 
@@ -242,11 +246,11 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 
 	ctxt->ldt_ents = 0;
 
-	BUG_ON((unsigned long)gdt->gdt & ~PAGE_MASK);
-	make_lowmem_page_readonly(gdt->gdt);
+	BUG_ON((unsigned long)gdt & ~PAGE_MASK);
+	make_lowmem_page_readonly(gdt);
 
-	ctxt->gdt_frames[0] = virt_to_mfn(gdt->gdt);
-	ctxt->gdt_ents      = ARRAY_SIZE(gdt->gdt);
+	ctxt->gdt_frames[0] = virt_to_mfn(gdt);
+	ctxt->gdt_ents      = GDT_ENTRIES;
 
 	ctxt->user_regs.cs = __KERNEL_CS;
 	ctxt->user_regs.esp = idle->thread.sp0 - sizeof(struct pt_regs);
@@ -254,9 +258,11 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 	ctxt->kernel_ss = __KERNEL_DS;
 	ctxt->kernel_sp = idle->thread.sp0;
 
+#ifdef CONFIG_X86_32
 	ctxt->event_callback_cs     = __KERNEL_CS;
-	ctxt->event_callback_eip    = (unsigned long)xen_hypervisor_callback;
 	ctxt->failsafe_callback_cs  = __KERNEL_CS;
+#endif
+	ctxt->event_callback_eip    = (unsigned long)xen_hypervisor_callback;
 	ctxt->failsafe_callback_eip = (unsigned long)xen_failsafe_callback;
 
 	per_cpu(xen_cr3, cpu) = __pa(swapper_pg_dir);
@@ -269,21 +275,33 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 	return 0;
 }
 
-int __cpuinit xen_cpu_up(unsigned int cpu)
+static int __cpuinit xen_cpu_up(unsigned int cpu)
 {
 	struct task_struct *idle = idle_task(cpu);
 	int rc;
 
-#if 0
-	rc = cpu_up_check(cpu);
-	if (rc)
-		return rc;
+#ifdef CONFIG_X86_64
+	/* Allocate node local memory for AP pdas */
+	WARN_ON(cpu == 0);
+	if (cpu > 0) {
+		rc = get_local_pda(cpu);
+		if (rc)
+			return rc;
+	}
 #endif
 
+#ifdef CONFIG_X86_32
 	init_gdt(cpu);
 	per_cpu(current_task, cpu) = idle;
 	irq_ctx_init(cpu);
+#else
+	cpu_pda(cpu)->pcurrent = idle;
+	clear_tsk_thread_flag(idle, TIF_FORK);
+#endif
 	xen_setup_timer(cpu);
+	xen_init_lock_cpu(cpu);
+
+	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
 
 	/* make sure interrupts start blocked */
 	per_cpu(xen_vcpu, cpu)->evtchn_upcall_mask = 1;
@@ -299,23 +317,75 @@ int __cpuinit xen_cpu_up(unsigned int cpu)
 	if (rc)
 		return rc;
 
-	smp_store_cpu_info(cpu);
-	set_cpu_sibling_map(cpu);
-	/* This must be done before setting cpu_online_map */
-	wmb();
-
-	cpu_set(cpu, cpu_online_map);
-
 	rc = HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL);
 	BUG_ON(rc);
 
+	while(per_cpu(cpu_state, cpu) != CPU_ONLINE) {
+		HYPERVISOR_sched_op(SCHEDOP_yield, 0);
+		barrier();
+	}
+
+	return 0;
+}
+
+static void xen_smp_cpus_done(unsigned int max_cpus)
+{
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static int xen_cpu_disable(void)
+{
+	unsigned int cpu = smp_processor_id();
+	if (cpu == 0)
+		return -EBUSY;
+
+	cpu_disable_common();
+
+	load_cr3(swapper_pg_dir);
 	return 0;
 }
 
-void xen_smp_cpus_done(unsigned int max_cpus)
+static void xen_cpu_die(unsigned int cpu)
+{
+	while (HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL)) {
+		current->state = TASK_UNINTERRUPTIBLE;
+		schedule_timeout(HZ/10);
+	}
+	unbind_from_irqhandler(per_cpu(resched_irq, cpu), NULL);
+	unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
+	unbind_from_irqhandler(per_cpu(debug_irq, cpu), NULL);
+	unbind_from_irqhandler(per_cpu(callfuncsingle_irq, cpu), NULL);
+	xen_uninit_lock_cpu(cpu);
+	xen_teardown_timer(cpu);
+
+	if (num_online_cpus() == 1)
+		alternatives_smp_switch(0);
+}
+
+static void xen_play_dead(void)
+{
+	play_dead_common();
+	HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
+	cpu_bringup();
+}
+
+#else /* !CONFIG_HOTPLUG_CPU */
+static int xen_cpu_disable(void)
 {
+	return -ENOSYS;
 }
 
+static void xen_cpu_die(unsigned int cpu)
+{
+	BUG();
+}
+
+static void xen_play_dead(void)
+{
+	BUG();
+}
+
+#endif
 static void stop_self(void *v)
 {
 	int cpu = smp_processor_id();
@@ -328,104 +398,94 @@ static void stop_self(void *v)
 	BUG();
 }
 
-void xen_smp_send_stop(void)
+static void xen_smp_send_stop(void)
 {
-	smp_call_function(stop_self, NULL, 0, 0);
+	smp_call_function(stop_self, NULL, 0);
 }
 
-void xen_smp_send_reschedule(int cpu)
+static void xen_smp_send_reschedule(int cpu)
 {
 	xen_send_IPI_one(cpu, XEN_RESCHEDULE_VECTOR);
 }
 
-
 static void xen_send_IPI_mask(cpumask_t mask, enum ipi_vector vector)
 {
 	unsigned cpu;
 
 	cpus_and(mask, mask, cpu_online_map);
 
-	for_each_cpu_mask(cpu, mask)
+	for_each_cpu_mask_nr(cpu, mask)
 		xen_send_IPI_one(cpu, vector);
 }
 
+static void xen_smp_send_call_function_ipi(cpumask_t mask)
+{
+	int cpu;
+
+	xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
+
+	/* Make sure other vcpus get a chance to run if they need to. */
+	for_each_cpu_mask_nr(cpu, mask) {
+		if (xen_vcpu_stolen(cpu)) {
+			HYPERVISOR_sched_op(SCHEDOP_yield, 0);
+			break;
+		}
+	}
+}
+
+static void xen_smp_send_call_function_single_ipi(int cpu)
+{
+	xen_send_IPI_mask(cpumask_of_cpu(cpu), XEN_CALL_FUNCTION_SINGLE_VECTOR);
+}
+
 static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id)
 {
-	void (*func) (void *info) = call_data->func;
-	void *info = call_data->info;
-	int wait = call_data->wait;
-
-	/*
-	 * Notify initiating CPU that I've grabbed the data and am
-	 * about to execute the function
-	 */
-	mb();
-	atomic_inc(&call_data->started);
-	/*
-	 * At this point the info structure may be out of scope unless wait==1
-	 */
 	irq_enter();
-	(*func)(info);
+	generic_smp_call_function_interrupt();
+#ifdef CONFIG_X86_32
 	__get_cpu_var(irq_stat).irq_call_count++;
+#else
+	add_pda(irq_call_count, 1);
+#endif
 	irq_exit();
 
-	if (wait) {
-		mb();		/* commit everything before setting finished */
-		atomic_inc(&call_data->finished);
-	}
-
 	return IRQ_HANDLED;
 }
 
-int xen_smp_call_function_mask(cpumask_t mask, void (*func)(void *),
-			       void *info, int wait)
+static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id)
 {
-	struct call_data_struct data;
-	int cpus, cpu;
-	bool yield;
-
-	/* Holding any lock stops cpus from going down. */
-	spin_lock(&call_lock);
-
-	cpu_clear(smp_processor_id(), mask);
-
-	cpus = cpus_weight(mask);
-	if (!cpus) {
-		spin_unlock(&call_lock);
-		return 0;
-	}
-
-	/* Can deadlock when called with interrupts disabled */
-	WARN_ON(irqs_disabled());
-
-	data.func = func;
-	data.info = info;
-	atomic_set(&data.started, 0);
-	data.wait = wait;
-	if (wait)
-		atomic_set(&data.finished, 0);
-
-	call_data = &data;
-	mb();			/* write everything before IPI */
+	irq_enter();
+	generic_smp_call_function_single_interrupt();
+#ifdef CONFIG_X86_32
+	__get_cpu_var(irq_stat).irq_call_count++;
+#else
+	add_pda(irq_call_count, 1);
+#endif
+	irq_exit();
 
-	/* Send a message to other CPUs and wait for them to respond */
-	xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
+	return IRQ_HANDLED;
+}
 
-	/* Make sure other vcpus get a chance to run if they need to. */
-	yield = false;
-	for_each_cpu_mask(cpu, mask)
-		if (xen_vcpu_stolen(cpu))
-			yield = true;
+static const struct smp_ops xen_smp_ops __initdata = {
+	.smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu,
+	.smp_prepare_cpus = xen_smp_prepare_cpus,
+	.smp_cpus_done = xen_smp_cpus_done,
 
-	if (yield)
-		HYPERVISOR_sched_op(SCHEDOP_yield, 0);
+	.cpu_up = xen_cpu_up,
+	.cpu_die = xen_cpu_die,
+	.cpu_disable = xen_cpu_disable,
+	.play_dead = xen_play_dead,
 
-	/* Wait for response */
-	while (atomic_read(&data.started) != cpus ||
-	       (wait && atomic_read(&data.finished) != cpus))
-		cpu_relax();
+	.smp_send_stop = xen_smp_send_stop,
+	.smp_send_reschedule = xen_smp_send_reschedule,
 
-	spin_unlock(&call_lock);
+	.send_call_func_ipi = xen_smp_send_call_function_ipi,
+	.send_call_func_single_ipi = xen_smp_send_call_function_single_ipi,
+};
 
-	return 0;
+void __init xen_smp_init(void)
+{
+	smp_ops = xen_smp_ops;
+	xen_fill_possible_map();
+	xen_init_spinlocks();
 }
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
new file mode 100644
index 00000000000..dd71e3a021c
--- /dev/null
+++ b/arch/x86/xen/spinlock.c
@@ -0,0 +1,428 @@
+/*
+ * Split spinlock implementation out into its own file, so it can be
+ * compiled in a FTRACE-compatible way.
+ */
+#include <linux/kernel_stat.h>
+#include <linux/spinlock.h>
+#include <linux/debugfs.h>
+#include <linux/log2.h>
+
+#include <asm/paravirt.h>
+
+#include <xen/interface/xen.h>
+#include <xen/events.h>
+
+#include "xen-ops.h"
+#include "debugfs.h"
+
+#ifdef CONFIG_XEN_DEBUG_FS
+static struct xen_spinlock_stats
+{
+	u64 taken;
+	u32 taken_slow;
+	u32 taken_slow_nested;
+	u32 taken_slow_pickup;
+	u32 taken_slow_spurious;
+	u32 taken_slow_irqenable;
+
+	u64 released;
+	u32 released_slow;
+	u32 released_slow_kicked;
+
+#define HISTO_BUCKETS	30
+	u32 histo_spin_total[HISTO_BUCKETS+1];
+	u32 histo_spin_spinning[HISTO_BUCKETS+1];
+	u32 histo_spin_blocked[HISTO_BUCKETS+1];
+
+	u64 time_total;
+	u64 time_spinning;
+	u64 time_blocked;
+} spinlock_stats;
+
+static u8 zero_stats;
+
+static unsigned lock_timeout = 1 << 10;
+#define TIMEOUT lock_timeout
+
+static inline void check_zero(void)
+{
+	if (unlikely(zero_stats)) {
+		memset(&spinlock_stats, 0, sizeof(spinlock_stats));
+		zero_stats = 0;
+	}
+}
+
+#define ADD_STATS(elem, val)			\
+	do { check_zero(); spinlock_stats.elem += (val); } while(0)
+
+static inline u64 spin_time_start(void)
+{
+	return xen_clocksource_read();
+}
+
+static void __spin_time_accum(u64 delta, u32 *array)
+{
+	unsigned index = ilog2(delta);
+
+	check_zero();
+
+	if (index < HISTO_BUCKETS)
+		array[index]++;
+	else
+		array[HISTO_BUCKETS]++;
+}
+
+static inline void spin_time_accum_spinning(u64 start)
+{
+	u32 delta = xen_clocksource_read() - start;
+
+	__spin_time_accum(delta, spinlock_stats.histo_spin_spinning);
+	spinlock_stats.time_spinning += delta;
+}
+
+static inline void spin_time_accum_total(u64 start)
+{
+	u32 delta = xen_clocksource_read() - start;
+
+	__spin_time_accum(delta, spinlock_stats.histo_spin_total);
+	spinlock_stats.time_total += delta;
+}
+
+static inline void spin_time_accum_blocked(u64 start)
+{
+	u32 delta = xen_clocksource_read() - start;
+
+	__spin_time_accum(delta, spinlock_stats.histo_spin_blocked);
+	spinlock_stats.time_blocked += delta;
+}
+#else  /* !CONFIG_XEN_DEBUG_FS */
+#define TIMEOUT			(1 << 10)
+#define ADD_STATS(elem, val)	do { (void)(val); } while(0)
+
+static inline u64 spin_time_start(void)
+{
+	return 0;
+}
+
+static inline void spin_time_accum_total(u64 start)
+{
+}
+static inline void spin_time_accum_spinning(u64 start)
+{
+}
+static inline void spin_time_accum_blocked(u64 start)
+{
+}
+#endif  /* CONFIG_XEN_DEBUG_FS */
+
+struct xen_spinlock {
+	unsigned char lock;		/* 0 -> free; 1 -> locked */
+	unsigned short spinners;	/* count of waiting cpus */
+};
+
+static int xen_spin_is_locked(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+
+	return xl->lock != 0;
+}
+
+static int xen_spin_is_contended(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+
+	/* Not strictly true; this is only the count of contended
+	   lock-takers entering the slow path. */
+	return xl->spinners != 0;
+}
+
+static int xen_spin_trylock(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+	u8 old = 1;
+
+	asm("xchgb %b0,%1"
+	    : "+q" (old), "+m" (xl->lock) : : "memory");
+
+	return old == 0;
+}
+
+static DEFINE_PER_CPU(int, lock_kicker_irq) = -1;
+static DEFINE_PER_CPU(struct xen_spinlock *, lock_spinners);
+
+/*
+ * Mark a cpu as interested in a lock.  Returns the CPU's previous
+ * lock of interest, in case we got preempted by an interrupt.
+ */
+static inline struct xen_spinlock *spinning_lock(struct xen_spinlock *xl)
+{
+	struct xen_spinlock *prev;
+
+	prev = __get_cpu_var(lock_spinners);
+	__get_cpu_var(lock_spinners) = xl;
+
+	wmb();			/* set lock of interest before count */
+
+	asm(LOCK_PREFIX " incw %0"
+	    : "+m" (xl->spinners) : : "memory");
+
+	return prev;
+}
+
+/*
+ * Mark a cpu as no longer interested in a lock.  Restores previous
+ * lock of interest (NULL for none).
+ */
+static inline void unspinning_lock(struct xen_spinlock *xl, struct xen_spinlock *prev)
+{
+	asm(LOCK_PREFIX " decw %0"
+	    : "+m" (xl->spinners) : : "memory");
+	wmb();			/* decrement count before restoring lock */
+	__get_cpu_var(lock_spinners) = prev;
+}
+
+static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enable)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+	struct xen_spinlock *prev;
+	int irq = __get_cpu_var(lock_kicker_irq);
+	int ret;
+	unsigned long flags;
+	u64 start;
+
+	/* If kicker interrupts not initialized yet, just spin */
+	if (irq == -1)
+		return 0;
+
+	start = spin_time_start();
+
+	/* announce we're spinning */
+	prev = spinning_lock(xl);
+
+	flags = __raw_local_save_flags();
+	if (irq_enable) {
+		ADD_STATS(taken_slow_irqenable, 1);
+		raw_local_irq_enable();
+	}
+
+	ADD_STATS(taken_slow, 1);
+	ADD_STATS(taken_slow_nested, prev != NULL);
+
+	do {
+		/* clear pending */
+		xen_clear_irq_pending(irq);
+
+		/* check again make sure it didn't become free while
+		   we weren't looking  */
+		ret = xen_spin_trylock(lock);
+		if (ret) {
+			ADD_STATS(taken_slow_pickup, 1);
+
+			/*
+			 * If we interrupted another spinlock while it
+			 * was blocking, make sure it doesn't block
+			 * without rechecking the lock.
+			 */
+			if (prev != NULL)
+				xen_set_irq_pending(irq);
+			goto out;
+		}
+
+		/*
+		 * Block until irq becomes pending.  If we're
+		 * interrupted at this point (after the trylock but
+		 * before entering the block), then the nested lock
+		 * handler guarantees that the irq will be left
+		 * pending if there's any chance the lock became free;
+		 * xen_poll_irq() returns immediately if the irq is
+		 * pending.
+		 */
+		xen_poll_irq(irq);
+		ADD_STATS(taken_slow_spurious, !xen_test_irq_pending(irq));
+	} while (!xen_test_irq_pending(irq)); /* check for spurious wakeups */
+
+	kstat_this_cpu.irqs[irq]++;
+
+out:
+	raw_local_irq_restore(flags);
+	unspinning_lock(xl, prev);
+	spin_time_accum_blocked(start);
+
+	return ret;
+}
+
+static inline void __xen_spin_lock(struct raw_spinlock *lock, bool irq_enable)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+	unsigned timeout;
+	u8 oldval;
+	u64 start_spin;
+
+	ADD_STATS(taken, 1);
+
+	start_spin = spin_time_start();
+
+	do {
+		u64 start_spin_fast = spin_time_start();
+
+		timeout = TIMEOUT;
+
+		asm("1: xchgb %1,%0\n"
+		    "   testb %1,%1\n"
+		    "   jz 3f\n"
+		    "2: rep;nop\n"
+		    "   cmpb $0,%0\n"
+		    "   je 1b\n"
+		    "   dec %2\n"
+		    "   jnz 2b\n"
+		    "3:\n"
+		    : "+m" (xl->lock), "=q" (oldval), "+r" (timeout)
+		    : "1" (1)
+		    : "memory");
+
+		spin_time_accum_spinning(start_spin_fast);
+
+	} while (unlikely(oldval != 0 &&
+			  (TIMEOUT == ~0 || !xen_spin_lock_slow(lock, irq_enable))));
+
+	spin_time_accum_total(start_spin);
+}
+
+static void xen_spin_lock(struct raw_spinlock *lock)
+{
+	__xen_spin_lock(lock, false);
+}
+
+static void xen_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags)
+{
+	__xen_spin_lock(lock, !raw_irqs_disabled_flags(flags));
+}
+
+static noinline void xen_spin_unlock_slow(struct xen_spinlock *xl)
+{
+	int cpu;
+
+	ADD_STATS(released_slow, 1);
+
+	for_each_online_cpu(cpu) {
+		/* XXX should mix up next cpu selection */
+		if (per_cpu(lock_spinners, cpu) == xl) {
+			ADD_STATS(released_slow_kicked, 1);
+			xen_send_IPI_one(cpu, XEN_SPIN_UNLOCK_VECTOR);
+			break;
+		}
+	}
+}
+
+static void xen_spin_unlock(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+
+	ADD_STATS(released, 1);
+
+	smp_wmb();		/* make sure no writes get moved after unlock */
+	xl->lock = 0;		/* release lock */
+
+	/* make sure unlock happens before kick */
+	barrier();
+
+	if (unlikely(xl->spinners))
+		xen_spin_unlock_slow(xl);
+}
+
+static irqreturn_t dummy_handler(int irq, void *dev_id)
+{
+	BUG();
+	return IRQ_HANDLED;
+}
+
+void __cpuinit xen_init_lock_cpu(int cpu)
+{
+	int irq;
+	const char *name;
+
+	name = kasprintf(GFP_KERNEL, "spinlock%d", cpu);
+	irq = bind_ipi_to_irqhandler(XEN_SPIN_UNLOCK_VECTOR,
+				     cpu,
+				     dummy_handler,
+				     IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				     name,
+				     NULL);
+
+	if (irq >= 0) {
+		disable_irq(irq); /* make sure it's never delivered */
+		per_cpu(lock_kicker_irq, cpu) = irq;
+	}
+
+	printk("cpu %d spinlock event irq %d\n", cpu, irq);
+}
+
+void xen_uninit_lock_cpu(int cpu)
+{
+	unbind_from_irqhandler(per_cpu(lock_kicker_irq, cpu), NULL);
+}
+
+void __init xen_init_spinlocks(void)
+{
+	pv_lock_ops.spin_is_locked = xen_spin_is_locked;
+	pv_lock_ops.spin_is_contended = xen_spin_is_contended;
+	pv_lock_ops.spin_lock = xen_spin_lock;
+	pv_lock_ops.spin_lock_flags = xen_spin_lock_flags;
+	pv_lock_ops.spin_trylock = xen_spin_trylock;
+	pv_lock_ops.spin_unlock = xen_spin_unlock;
+}
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct dentry *d_spin_debug;
+
+static int __init xen_spinlock_debugfs(void)
+{
+	struct dentry *d_xen = xen_init_debugfs();
+
+	if (d_xen == NULL)
+		return -ENOMEM;
+
+	d_spin_debug = debugfs_create_dir("spinlocks", d_xen);
+
+	debugfs_create_u8("zero_stats", 0644, d_spin_debug, &zero_stats);
+
+	debugfs_create_u32("timeout", 0644, d_spin_debug, &lock_timeout);
+
+	debugfs_create_u64("taken", 0444, d_spin_debug, &spinlock_stats.taken);
+	debugfs_create_u32("taken_slow", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow);
+	debugfs_create_u32("taken_slow_nested", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_nested);
+	debugfs_create_u32("taken_slow_pickup", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_pickup);
+	debugfs_create_u32("taken_slow_spurious", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_spurious);
+	debugfs_create_u32("taken_slow_irqenable", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_irqenable);
+
+	debugfs_create_u64("released", 0444, d_spin_debug, &spinlock_stats.released);
+	debugfs_create_u32("released_slow", 0444, d_spin_debug,
+			   &spinlock_stats.released_slow);
+	debugfs_create_u32("released_slow_kicked", 0444, d_spin_debug,
+			   &spinlock_stats.released_slow_kicked);
+
+	debugfs_create_u64("time_spinning", 0444, d_spin_debug,
+			   &spinlock_stats.time_spinning);
+	debugfs_create_u64("time_blocked", 0444, d_spin_debug,
+			   &spinlock_stats.time_blocked);
+	debugfs_create_u64("time_total", 0444, d_spin_debug,
+			   &spinlock_stats.time_total);
+
+	xen_debugfs_create_u32_array("histo_total", 0444, d_spin_debug,
+				     spinlock_stats.histo_spin_total, HISTO_BUCKETS + 1);
+	xen_debugfs_create_u32_array("histo_spinning", 0444, d_spin_debug,
+				     spinlock_stats.histo_spin_spinning, HISTO_BUCKETS + 1);
+	xen_debugfs_create_u32_array("histo_blocked", 0444, d_spin_debug,
+				     spinlock_stats.histo_spin_blocked, HISTO_BUCKETS + 1);
+
+	return 0;
+}
+fs_initcall(xen_spinlock_debugfs);
+
+#endif	/* CONFIG_XEN_DEBUG_FS */
diff --git a/arch/x86/xen/suspend.c b/arch/x86/xen/suspend.c
new file mode 100644
index 00000000000..2a234db5949
--- /dev/null
+++ b/arch/x86/xen/suspend.c
@@ -0,0 +1,48 @@
+#include <linux/types.h>
+
+#include <xen/interface/xen.h>
+#include <xen/grant_table.h>
+#include <xen/events.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+#include "xen-ops.h"
+#include "mmu.h"
+
+void xen_pre_suspend(void)
+{
+	xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn);
+	xen_start_info->console.domU.mfn =
+		mfn_to_pfn(xen_start_info->console.domU.mfn);
+
+	BUG_ON(!irqs_disabled());
+
+	HYPERVISOR_shared_info = &xen_dummy_shared_info;
+	if (HYPERVISOR_update_va_mapping(fix_to_virt(FIX_PARAVIRT_BOOTMAP),
+					 __pte_ma(0), 0))
+		BUG();
+}
+
+void xen_post_suspend(int suspend_cancelled)
+{
+	xen_setup_shared_info();
+
+	if (suspend_cancelled) {
+		xen_start_info->store_mfn =
+			pfn_to_mfn(xen_start_info->store_mfn);
+		xen_start_info->console.domU.mfn =
+			pfn_to_mfn(xen_start_info->console.domU.mfn);
+	} else {
+#ifdef CONFIG_SMP
+		xen_cpu_initialized_map = cpu_online_map;
+#endif
+		xen_vcpu_restore();
+	}
+
+}
+
+void xen_arch_resume(void)
+{
+	/* nothing */
+}
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index 41e217503c9..004ba86326a 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -30,8 +30,6 @@
 #define TIMER_SLOP	100000
 #define NS_PER_TICK	(1000000000LL / HZ)
 
-static cycle_t xen_clocksource_read(void);
-
 /* runstate info updated by Xen */
 static DEFINE_PER_CPU(struct vcpu_runstate_info, runstate);
 
@@ -197,8 +195,8 @@ unsigned long long xen_sched_clock(void)
 }
 
 
-/* Get the CPU speed from Xen */
-unsigned long xen_cpu_khz(void)
+/* Get the TSC speed from Xen */
+unsigned long xen_tsc_khz(void)
 {
 	u64 xen_khz = 1000000ULL << 32;
 	const struct pvclock_vcpu_time_info *info =
@@ -213,7 +211,7 @@ unsigned long xen_cpu_khz(void)
 	return xen_khz;
 }
 
-static cycle_t xen_clocksource_read(void)
+cycle_t xen_clocksource_read(void)
 {
         struct pvclock_vcpu_time_info *src;
 	cycle_t ret;
@@ -452,6 +450,14 @@ void xen_setup_timer(int cpu)
 	setup_runstate_info(cpu);
 }
 
+void xen_teardown_timer(int cpu)
+{
+	struct clock_event_device *evt;
+	BUG_ON(cpu == 0);
+	evt = &per_cpu(xen_clock_events, cpu);
+	unbind_from_irqhandler(evt->irq, NULL);
+}
+
 void xen_setup_cpu_clockevents(void)
 {
 	BUG_ON(preemptible());
@@ -459,6 +465,19 @@ void xen_setup_cpu_clockevents(void)
 	clockevents_register_device(&__get_cpu_var(xen_clock_events));
 }
 
+void xen_timer_resume(void)
+{
+	int cpu;
+
+	if (xen_clockevent != &xen_vcpuop_clockevent)
+		return;
+
+	for_each_online_cpu(cpu) {
+		if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL))
+			BUG();
+	}
+}
+
 __init void xen_time_init(void)
 {
 	int cpu = smp_processor_id();
diff --git a/arch/x86/xen/xen-asm.S b/arch/x86/xen/xen-asm_32.S
index 2497a30f41d..42786f59d9c 100644
--- a/arch/x86/xen/xen-asm.S
+++ b/arch/x86/xen/xen-asm_32.S
@@ -298,7 +298,7 @@ check_events:
 	push %eax
 	push %ecx
 	push %edx
-	call force_evtchn_callback
+	call xen_force_evtchn_callback
 	pop %edx
 	pop %ecx
 	pop %eax
diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S
new file mode 100644
index 00000000000..05794c566e8
--- /dev/null
+++ b/arch/x86/xen/xen-asm_64.S
@@ -0,0 +1,285 @@
+/*
+	Asm versions of Xen pv-ops, suitable for either direct use or inlining.
+	The inline versions are the same as the direct-use versions, with the
+	pre- and post-amble chopped off.
+
+	This code is encoded for size rather than absolute efficiency,
+	with a view to being able to inline as much as possible.
+
+	We only bother with direct forms (ie, vcpu in pda) of the operations
+	here; the indirect forms are better handled in C, since they're
+	generally too large to inline anyway.
+ */
+
+#include <linux/linkage.h>
+
+#include <asm/asm-offsets.h>
+#include <asm/processor-flags.h>
+#include <asm/errno.h>
+#include <asm/segment.h>
+
+#include <xen/interface/xen.h>
+
+#define RELOC(x, v)	.globl x##_reloc; x##_reloc=v
+#define ENDPATCH(x)	.globl x##_end; x##_end=.
+
+/* Pseudo-flag used for virtual NMI, which we don't implement yet */
+#define XEN_EFLAGS_NMI	0x80000000
+
+#if 1
+/*
+	x86-64 does not yet support direct access to percpu variables
+	via a segment override, so we just need to make sure this code
+	never gets used
+ */
+#define BUG			ud2a
+#define PER_CPU_VAR(var, off)	0xdeadbeef
+#endif
+
+/*
+	Enable events.  This clears the event mask and tests the pending
+	event status with one and operation.  If there are pending
+	events, then enter the hypervisor to get them handled.
+ */
+ENTRY(xen_irq_enable_direct)
+	BUG
+
+	/* Unmask events */
+	movb $0, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
+
+	/* Preempt here doesn't matter because that will deal with
+	   any pending interrupts.  The pending check may end up being
+	   run on the wrong CPU, but that doesn't hurt. */
+
+	/* Test for pending */
+	testb $0xff, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_pending)
+	jz 1f
+
+2:	call check_events
+1:
+ENDPATCH(xen_irq_enable_direct)
+	ret
+	ENDPROC(xen_irq_enable_direct)
+	RELOC(xen_irq_enable_direct, 2b+1)
+
+/*
+	Disabling events is simply a matter of making the event mask
+	non-zero.
+ */
+ENTRY(xen_irq_disable_direct)
+	BUG
+
+	movb $1, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
+ENDPATCH(xen_irq_disable_direct)
+	ret
+	ENDPROC(xen_irq_disable_direct)
+	RELOC(xen_irq_disable_direct, 0)
+
+/*
+	(xen_)save_fl is used to get the current interrupt enable status.
+	Callers expect the status to be in X86_EFLAGS_IF, and other bits
+	may be set in the return value.  We take advantage of this by
+	making sure that X86_EFLAGS_IF has the right value (and other bits
+	in that byte are 0), but other bits in the return value are
+	undefined.  We need to toggle the state of the bit, because
+	Xen and x86 use opposite senses (mask vs enable).
+ */
+ENTRY(xen_save_fl_direct)
+	BUG
+
+	testb $0xff, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
+	setz %ah
+	addb %ah,%ah
+ENDPATCH(xen_save_fl_direct)
+	ret
+	ENDPROC(xen_save_fl_direct)
+	RELOC(xen_save_fl_direct, 0)
+
+/*
+	In principle the caller should be passing us a value return
+	from xen_save_fl_direct, but for robustness sake we test only
+	the X86_EFLAGS_IF flag rather than the whole byte. After
+	setting the interrupt mask state, it checks for unmasked
+	pending events and enters the hypervisor to get them delivered
+	if so.
+ */
+ENTRY(xen_restore_fl_direct)
+	BUG
+
+	testb $X86_EFLAGS_IF>>8, %ah
+	setz PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
+	/* Preempt here doesn't matter because that will deal with
+	   any pending interrupts.  The pending check may end up being
+	   run on the wrong CPU, but that doesn't hurt. */
+
+	/* check for unmasked and pending */
+	cmpw $0x0001, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_pending)
+	jz 1f
+2:	call check_events
+1:
+ENDPATCH(xen_restore_fl_direct)
+	ret
+	ENDPROC(xen_restore_fl_direct)
+	RELOC(xen_restore_fl_direct, 2b+1)
+
+
+/*
+	Force an event check by making a hypercall,
+	but preserve regs before making the call.
+ */
+check_events:
+	push %rax
+	push %rcx
+	push %rdx
+	push %rsi
+	push %rdi
+	push %r8
+	push %r9
+	push %r10
+	push %r11
+	call xen_force_evtchn_callback
+	pop %r11
+	pop %r10
+	pop %r9
+	pop %r8
+	pop %rdi
+	pop %rsi
+	pop %rdx
+	pop %rcx
+	pop %rax
+	ret
+
+ENTRY(xen_adjust_exception_frame)
+	mov 8+0(%rsp),%rcx
+	mov 8+8(%rsp),%r11
+	ret $16
+
+hypercall_iret = hypercall_page + __HYPERVISOR_iret * 32
+/*
+	Xen64 iret frame:
+
+	ss
+	rsp
+	rflags
+	cs
+	rip		<-- standard iret frame
+
+	flags
+
+	rcx		}
+	r11		}<-- pushed by hypercall page
+rsp ->	rax		}
+ */
+ENTRY(xen_iret)
+	pushq $0
+1:	jmp hypercall_iret
+ENDPATCH(xen_iret)
+RELOC(xen_iret, 1b+1)
+
+/*
+	sysexit is not used for 64-bit processes, so it's
+	only ever used to return to 32-bit compat userspace.
+ */
+ENTRY(xen_sysexit)
+	pushq $__USER32_DS
+	pushq %rcx
+	pushq $X86_EFLAGS_IF
+	pushq $__USER32_CS
+	pushq %rdx
+
+	pushq $0
+1:	jmp hypercall_iret
+ENDPATCH(xen_sysexit)
+RELOC(xen_sysexit, 1b+1)
+
+ENTRY(xen_sysret64)
+	/* We're already on the usermode stack at this point, but still
+	   with the kernel gs, so we can easily switch back */
+	movq %rsp, %gs:pda_oldrsp
+	movq %gs:pda_kernelstack,%rsp
+
+	pushq $__USER_DS
+	pushq %gs:pda_oldrsp
+	pushq %r11
+	pushq $__USER_CS
+	pushq %rcx
+
+	pushq $VGCF_in_syscall
+1:	jmp hypercall_iret
+ENDPATCH(xen_sysret64)
+RELOC(xen_sysret64, 1b+1)
+
+ENTRY(xen_sysret32)
+	/* We're already on the usermode stack at this point, but still
+	   with the kernel gs, so we can easily switch back */
+	movq %rsp, %gs:pda_oldrsp
+	movq %gs:pda_kernelstack, %rsp
+
+	pushq $__USER32_DS
+	pushq %gs:pda_oldrsp
+	pushq %r11
+	pushq $__USER32_CS
+	pushq %rcx
+
+	pushq $VGCF_in_syscall
+1:	jmp hypercall_iret
+ENDPATCH(xen_sysret32)
+RELOC(xen_sysret32, 1b+1)
+
+/*
+	Xen handles syscall callbacks much like ordinary exceptions,
+	which means we have:
+	 - kernel gs
+	 - kernel rsp
+	 - an iret-like stack frame on the stack (including rcx and r11):
+		ss
+		rsp
+		rflags
+		cs
+		rip
+		r11
+	rsp->	rcx
+
+	In all the entrypoints, we undo all that to make it look
+	like a CPU-generated syscall/sysenter and jump to the normal
+	entrypoint.
+ */
+
+.macro undo_xen_syscall
+	mov 0*8(%rsp),%rcx
+	mov 1*8(%rsp),%r11
+	mov 5*8(%rsp),%rsp
+.endm
+
+/* Normal 64-bit system call target */
+ENTRY(xen_syscall_target)
+	undo_xen_syscall
+	jmp system_call_after_swapgs
+ENDPROC(xen_syscall_target)
+
+#ifdef CONFIG_IA32_EMULATION
+
+/* 32-bit compat syscall target */
+ENTRY(xen_syscall32_target)
+	undo_xen_syscall
+	jmp ia32_cstar_target
+ENDPROC(xen_syscall32_target)
+
+/* 32-bit compat sysenter target */
+ENTRY(xen_sysenter_target)
+	undo_xen_syscall
+	jmp ia32_sysenter_target
+ENDPROC(xen_sysenter_target)
+
+#else /* !CONFIG_IA32_EMULATION */
+
+ENTRY(xen_syscall32_target)
+ENTRY(xen_sysenter_target)
+	lea 16(%rsp), %rsp	/* strip %rcx,%r11 */
+	mov $-ENOSYS, %rax
+	pushq $VGCF_in_syscall
+	jmp hypercall_iret
+ENDPROC(xen_syscall32_target)
+ENDPROC(xen_sysenter_target)
+
+#endif	/* CONFIG_IA32_EMULATION */
diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S
index 6ec3b4f7719..63d49a523ed 100644
--- a/arch/x86/xen/xen-head.S
+++ b/arch/x86/xen/xen-head.S
@@ -5,14 +5,24 @@
 
 #include <linux/elfnote.h>
 #include <linux/init.h>
+
 #include <asm/boot.h>
+#include <asm/asm.h>
+#include <asm/page.h>
+
 #include <xen/interface/elfnote.h>
+#include <asm/xen/interface.h>
 
 	__INIT
 ENTRY(startup_xen)
-	movl %esi,xen_start_info
 	cld
-	movl $(init_thread_union+THREAD_SIZE),%esp
+#ifdef CONFIG_X86_32
+	mov %esi,xen_start_info
+	mov $init_thread_union+THREAD_SIZE,%esp
+#else
+	mov %rsi,xen_start_info
+	mov $init_thread_union+THREAD_SIZE,%rsp
+#endif
 	jmp xen_start_kernel
 
 	__FINIT
@@ -20,17 +30,26 @@ ENTRY(startup_xen)
 .pushsection .text
 	.align PAGE_SIZE_asm
 ENTRY(hypercall_page)
-	.skip 0x1000
+	.skip PAGE_SIZE_asm
 .popsection
 
 	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS,       .asciz "linux")
 	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION,  .asciz "2.6")
 	ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION,    .asciz "xen-3.0")
-	ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE,      .long  __PAGE_OFFSET)
-	ELFNOTE(Xen, XEN_ELFNOTE_ENTRY,          .long  startup_xen)
-	ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .long  hypercall_page)
+#ifdef CONFIG_X86_32
+	ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE,      _ASM_PTR __PAGE_OFFSET)
+#else
+	ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE,      _ASM_PTR __START_KERNEL_map)
+#endif
+	ELFNOTE(Xen, XEN_ELFNOTE_ENTRY,          _ASM_PTR startup_xen)
+	ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, _ASM_PTR hypercall_page)
 	ELFNOTE(Xen, XEN_ELFNOTE_FEATURES,       .asciz "!writable_page_tables|pae_pgdir_above_4gb")
 	ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE,       .asciz "yes")
 	ELFNOTE(Xen, XEN_ELFNOTE_LOADER,         .asciz "generic")
+	ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID,
+		.quad _PAGE_PRESENT; .quad _PAGE_PRESENT)
+	ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long 1)
+	ELFNOTE(Xen, XEN_ELFNOTE_HV_START_LOW,   _ASM_PTR __HYPERVISOR_VIRT_START)
+	ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET,   _ASM_PTR 0)
 
 #endif /*CONFIG_XEN */
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index f1063ae0803..d7422dc2a55 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -2,6 +2,7 @@
 #define XEN_OPS_H
 
 #include <linux/init.h>
+#include <linux/clocksource.h>
 #include <linux/irqreturn.h>
 #include <xen/xen-ops.h>
 
@@ -9,22 +10,34 @@
 extern const char xen_hypervisor_callback[];
 extern const char xen_failsafe_callback[];
 
+struct trap_info;
 void xen_copy_trap_info(struct trap_info *traps);
 
 DECLARE_PER_CPU(unsigned long, xen_cr3);
 DECLARE_PER_CPU(unsigned long, xen_current_cr3);
 
 extern struct start_info *xen_start_info;
+extern struct shared_info xen_dummy_shared_info;
 extern struct shared_info *HYPERVISOR_shared_info;
 
+void xen_setup_mfn_list_list(void);
+void xen_setup_shared_info(void);
+
 char * __init xen_memory_setup(void);
 void __init xen_arch_setup(void);
 void __init xen_init_IRQ(void);
 void xen_enable_sysenter(void);
+void xen_enable_syscall(void);
+void xen_vcpu_restore(void);
+
+void __init xen_build_dynamic_phys_to_machine(void);
 
+void xen_init_irq_ops(void);
 void xen_setup_timer(int cpu);
+void xen_teardown_timer(int cpu);
+cycle_t xen_clocksource_read(void);
 void xen_setup_cpu_clockevents(void);
-unsigned long xen_cpu_khz(void);
+unsigned long xen_tsc_khz(void);
 void __init xen_time_init(void);
 unsigned long xen_get_wallclock(void);
 int xen_set_wallclock(unsigned long time);
@@ -36,23 +49,19 @@ bool xen_vcpu_stolen(int vcpu);
 
 void xen_mark_init_mm_pinned(void);
 
-void __init xen_fill_possible_map(void);
-
 void __init xen_setup_vcpu_info_placement(void);
-void xen_smp_prepare_boot_cpu(void);
-void xen_smp_prepare_cpus(unsigned int max_cpus);
-int xen_cpu_up(unsigned int cpu);
-void xen_smp_cpus_done(unsigned int max_cpus);
 
-void xen_smp_send_stop(void);
-void xen_smp_send_reschedule(int cpu);
-int xen_smp_call_function (void (*func) (void *info), void *info, int nonatomic,
-			   int wait);
-int xen_smp_call_function_single(int cpu, void (*func) (void *info), void *info,
-				 int nonatomic, int wait);
+#ifdef CONFIG_SMP
+void xen_smp_init(void);
+
+void __init xen_init_spinlocks(void);
+__cpuinit void xen_init_lock_cpu(int cpu);
+void xen_uninit_lock_cpu(int cpu);
 
-int xen_smp_call_function_mask(cpumask_t mask, void (*func)(void *),
-			       void *info, int wait);
+extern cpumask_t xen_cpu_initialized_map;
+#else
+static inline void xen_smp_init(void) {}
+#endif
 
 
 /* Declare an asm function, along with symbols needed to make it
@@ -67,7 +76,11 @@ DECL_ASM(void, xen_irq_disable_direct, void);
 DECL_ASM(unsigned long, xen_save_fl_direct, void);
 DECL_ASM(void, xen_restore_fl_direct, unsigned long);
 
+/* These are not functions, and cannot be called normally */
 void xen_iret(void);
 void xen_sysexit(void);
+void xen_sysret32(void);
+void xen_sysret64(void);
+void xen_adjust_exception_frame(void);
 
 #endif /* XEN_OPS_H */