Merge branch 'linus' into x86/spinlocks

Done to prevent this failure of an Octopus merge:

 Added arch/arm/include/asm/byteorder.h in both, but differently.
 ERROR: Merge conflict in arch/arm/include/asm/byteorder.h
 Auto-merging include/asm-x86/spinlock.h
 ERROR: Merge conflict in include/asm-x86/spinlock.h
 fatal: merge program failed

71 files changed, 4073 insertions, 3068 deletions

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 27ef365e757..c2ac1b4515a 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -58,7 +58,6 @@ EXPORT_SYMBOL(acpi_disabled);
 #ifdef	CONFIG_X86_64
 
 #include <asm/proto.h>
-#include <asm/genapic.h>
 
 #else				/* X86 */
 
@@ -97,8 +96,6 @@ static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE;
 #warning ACPI uses CMPXCHG, i486 and later hardware
 #endif
 
-static int acpi_mcfg_64bit_base_addr __initdata = FALSE;
-
 /* --------------------------------------------------------------------------
                               Boot-time Configuration
    -------------------------------------------------------------------------- */
@@ -160,6 +157,8 @@ char *__init __acpi_map_table(unsigned long phys, unsigned long size)
 struct acpi_mcfg_allocation *pci_mmcfg_config;
 int pci_mmcfg_config_num;
 
+static int acpi_mcfg_64bit_base_addr __initdata = FALSE;
+
 static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg)
 {
 	if (!strcmp(mcfg->header.oem_id, "SGI"))
@@ -253,10 +252,8 @@ static void __cpuinit acpi_register_lapic(int id, u8 enabled)
 		return;
 	}
 
-#ifdef CONFIG_X86_32
 	if (boot_cpu_physical_apicid != -1U)
 		ver = apic_version[boot_cpu_physical_apicid];
-#endif
 
 	generic_processor_info(id, ver);
 }
@@ -776,10 +773,8 @@ static void __init acpi_register_lapic_address(unsigned long address)
 	set_fixmap_nocache(FIX_APIC_BASE, address);
 	if (boot_cpu_physical_apicid == -1U) {
 		boot_cpu_physical_apicid  = read_apic_id();
-#ifdef CONFIG_X86_32
 		apic_version[boot_cpu_physical_apicid] =
 			 GET_APIC_VERSION(apic_read(APIC_LVR));
-#endif
 	}
 }
 
@@ -1607,6 +1602,14 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = {
 	 */
 	{
 	 .callback = dmi_ignore_irq0_timer_override,
+	 .ident = "HP nx6115 laptop",
+	 .matches = {
+		     DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+		     DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq nx6115"),
+		     },
+	 },
+	{
+	 .callback = dmi_ignore_irq0_timer_override,
 	 .ident = "HP NX6125 laptop",
 	 .matches = {
 		     DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
@@ -1621,6 +1624,14 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = {
 		     DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq nx6325"),
 		     },
 	 },
+	{
+	 .callback = dmi_ignore_irq0_timer_override,
+	 .ident = "HP 6715b laptop",
+	 .matches = {
+		     DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+		     DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq 6715b"),
+		     },
+	 },
 	{}
 };
 
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 69b4d060b21..34e4d112b1e 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -33,6 +33,10 @@
 
 static DEFINE_RWLOCK(amd_iommu_devtable_lock);
 
+/* A list of preallocated protection domains */
+static LIST_HEAD(iommu_pd_list);
+static DEFINE_SPINLOCK(iommu_pd_list_lock);
+
 /*
  * general struct to manage commands send to an IOMMU
  */
@@ -51,6 +55,102 @@ static int iommu_has_npcache(struct amd_iommu *iommu)
 
 /****************************************************************************
  *
+ * Interrupt handling functions
+ *
+ ****************************************************************************/
+
+static void iommu_print_event(void *__evt)
+{
+	u32 *event = __evt;
+	int type  = (event[1] >> EVENT_TYPE_SHIFT)  & EVENT_TYPE_MASK;
+	int devid = (event[0] >> EVENT_DEVID_SHIFT) & EVENT_DEVID_MASK;
+	int domid = (event[1] >> EVENT_DOMID_SHIFT) & EVENT_DOMID_MASK;
+	int flags = (event[1] >> EVENT_FLAGS_SHIFT) & EVENT_FLAGS_MASK;
+	u64 address = (u64)(((u64)event[3]) << 32) | event[2];
+
+	printk(KERN_ERR "AMD IOMMU: Event logged [");
+
+	switch (type) {
+	case EVENT_TYPE_ILL_DEV:
+		printk("ILLEGAL_DEV_TABLE_ENTRY device=%02x:%02x.%x "
+		       "address=0x%016llx flags=0x%04x]\n",
+		       PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid),
+		       address, flags);
+		break;
+	case EVENT_TYPE_IO_FAULT:
+		printk("IO_PAGE_FAULT device=%02x:%02x.%x "
+		       "domain=0x%04x address=0x%016llx flags=0x%04x]\n",
+		       PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid),
+		       domid, address, flags);
+		break;
+	case EVENT_TYPE_DEV_TAB_ERR:
+		printk("DEV_TAB_HARDWARE_ERROR device=%02x:%02x.%x "
+		       "address=0x%016llx flags=0x%04x]\n",
+		       PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid),
+		       address, flags);
+		break;
+	case EVENT_TYPE_PAGE_TAB_ERR:
+		printk("PAGE_TAB_HARDWARE_ERROR device=%02x:%02x.%x "
+		       "domain=0x%04x address=0x%016llx flags=0x%04x]\n",
+		       PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid),
+		       domid, address, flags);
+		break;
+	case EVENT_TYPE_ILL_CMD:
+		printk("ILLEGAL_COMMAND_ERROR address=0x%016llx]\n", address);
+		break;
+	case EVENT_TYPE_CMD_HARD_ERR:
+		printk("COMMAND_HARDWARE_ERROR address=0x%016llx "
+		       "flags=0x%04x]\n", address, flags);
+		break;
+	case EVENT_TYPE_IOTLB_INV_TO:
+		printk("IOTLB_INV_TIMEOUT device=%02x:%02x.%x "
+		       "address=0x%016llx]\n",
+		       PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid),
+		       address);
+		break;
+	case EVENT_TYPE_INV_DEV_REQ:
+		printk("INVALID_DEVICE_REQUEST device=%02x:%02x.%x "
+		       "address=0x%016llx flags=0x%04x]\n",
+		       PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid),
+		       address, flags);
+		break;
+	default:
+		printk(KERN_ERR "UNKNOWN type=0x%02x]\n", type);
+	}
+}
+
+static void iommu_poll_events(struct amd_iommu *iommu)
+{
+	u32 head, tail;
+	unsigned long flags;
+
+	spin_lock_irqsave(&iommu->lock, flags);
+
+	head = readl(iommu->mmio_base + MMIO_EVT_HEAD_OFFSET);
+	tail = readl(iommu->mmio_base + MMIO_EVT_TAIL_OFFSET);
+
+	while (head != tail) {
+		iommu_print_event(iommu->evt_buf + head);
+		head = (head + EVENT_ENTRY_SIZE) % iommu->evt_buf_size;
+	}
+
+	writel(head, iommu->mmio_base + MMIO_EVT_HEAD_OFFSET);
+
+	spin_unlock_irqrestore(&iommu->lock, flags);
+}
+
+irqreturn_t amd_iommu_int_handler(int irq, void *data)
+{
+	struct amd_iommu *iommu;
+
+	list_for_each_entry(iommu, &amd_iommu_list, list)
+		iommu_poll_events(iommu);
+
+	return IRQ_HANDLED;
+}
+
+/****************************************************************************
+ *
  * IOMMU command queuing functions
  *
  ****************************************************************************/
@@ -101,10 +201,10 @@ static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd)
  */
 static int iommu_completion_wait(struct amd_iommu *iommu)
 {
-	int ret, ready = 0;
+	int ret = 0, ready = 0;
 	unsigned status = 0;
 	struct iommu_cmd cmd;
-	unsigned long i = 0;
+	unsigned long flags, i = 0;
 
 	memset(&cmd, 0, sizeof(cmd));
 	cmd.data[0] = CMD_COMPL_WAIT_INT_MASK;
@@ -112,10 +212,12 @@ static int iommu_completion_wait(struct amd_iommu *iommu)
 
 	iommu->need_sync = 0;
 
-	ret = iommu_queue_command(iommu, &cmd);
+	spin_lock_irqsave(&iommu->lock, flags);
+
+	ret = __iommu_queue_command(iommu, &cmd);
 
 	if (ret)
-		return ret;
+		goto out;
 
 	while (!ready && (i < EXIT_LOOP_COUNT)) {
 		++i;
@@ -130,6 +232,8 @@ static int iommu_completion_wait(struct amd_iommu *iommu)
 
 	if (unlikely((i == EXIT_LOOP_COUNT) && printk_ratelimit()))
 		printk(KERN_WARNING "AMD IOMMU: Completion wait loop failed\n");
+out:
+	spin_unlock_irqrestore(&iommu->lock, flags);
 
 	return 0;
 }
@@ -140,6 +244,7 @@ static int iommu_completion_wait(struct amd_iommu *iommu)
 static int iommu_queue_inv_dev_entry(struct amd_iommu *iommu, u16 devid)
 {
 	struct iommu_cmd cmd;
+	int ret;
 
 	BUG_ON(iommu == NULL);
 
@@ -147,9 +252,11 @@ static int iommu_queue_inv_dev_entry(struct amd_iommu *iommu, u16 devid)
 	CMD_SET_TYPE(&cmd, CMD_INV_DEV_ENTRY);
 	cmd.data[0] = devid;
 
+	ret = iommu_queue_command(iommu, &cmd);
+
 	iommu->need_sync = 1;
 
-	return iommu_queue_command(iommu, &cmd);
+	return ret;
 }
 
 /*
@@ -159,6 +266,7 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu,
 		u64 address, u16 domid, int pde, int s)
 {
 	struct iommu_cmd cmd;
+	int ret;
 
 	memset(&cmd, 0, sizeof(cmd));
 	address &= PAGE_MASK;
@@ -171,9 +279,11 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu,
 	if (pde) /* PDE bit - we wan't flush everything not only the PTEs */
 		cmd.data[2] |= CMD_INV_IOMMU_PAGES_PDE_MASK;
 
+	ret = iommu_queue_command(iommu, &cmd);
+
 	iommu->need_sync = 1;
 
-	return iommu_queue_command(iommu, &cmd);
+	return ret;
 }
 
 /*
@@ -203,6 +313,14 @@ static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid,
 	return 0;
 }
 
+/* Flush the whole IO/TLB for a given protection domain */
+static void iommu_flush_tlb(struct amd_iommu *iommu, u16 domid)
+{
+	u64 address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS;
+
+	iommu_queue_inv_iommu_pages(iommu, address, domid, 0, 1);
+}
+
 /****************************************************************************
  *
  * The functions below are used the create the page table mappings for
@@ -362,11 +480,6 @@ static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom,
  * efficient allocator.
  *
  ****************************************************************************/
-static unsigned long dma_mask_to_pages(unsigned long mask)
-{
-	return (mask >> PAGE_SHIFT) +
-		(PAGE_ALIGN(mask & ~PAGE_MASK) >> PAGE_SHIFT);
-}
 
 /*
  * The address allocator core function.
@@ -375,25 +488,31 @@ static unsigned long dma_mask_to_pages(unsigned long mask)
  */
 static unsigned long dma_ops_alloc_addresses(struct device *dev,
 					     struct dma_ops_domain *dom,
-					     unsigned int pages)
+					     unsigned int pages,
+					     unsigned long align_mask,
+					     u64 dma_mask)
 {
-	unsigned long limit = dma_mask_to_pages(*dev->dma_mask);
+	unsigned long limit;
 	unsigned long address;
-	unsigned long size = dom->aperture_size >> PAGE_SHIFT;
 	unsigned long boundary_size;
 
 	boundary_size = ALIGN(dma_get_seg_boundary(dev) + 1,
 			PAGE_SIZE) >> PAGE_SHIFT;
-	limit = limit < size ? limit : size;
+	limit = iommu_device_max_index(dom->aperture_size >> PAGE_SHIFT, 0,
+				       dma_mask >> PAGE_SHIFT);
 
-	if (dom->next_bit >= limit)
+	if (dom->next_bit >= limit) {
 		dom->next_bit = 0;
+		dom->need_flush = true;
+	}
 
 	address = iommu_area_alloc(dom->bitmap, limit, dom->next_bit, pages,
-			0 , boundary_size, 0);
-	if (address == -1)
+				   0 , boundary_size, align_mask);
+	if (address == -1) {
 		address = iommu_area_alloc(dom->bitmap, limit, 0, pages,
-				0, boundary_size, 0);
+				0, boundary_size, align_mask);
+		dom->need_flush = true;
+	}
 
 	if (likely(address != -1)) {
 		dom->next_bit = address + pages;
@@ -459,7 +578,7 @@ static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 	if (start_page + pages > last_page)
 		pages = last_page - start_page;
 
-	set_bit_string(dom->bitmap, start_page, pages);
+	iommu_area_reserve(dom->bitmap, start_page, pages);
 }
 
 static void dma_ops_free_pagetable(struct dma_ops_domain *dma_dom)
@@ -553,6 +672,9 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu,
 	dma_dom->bitmap[0] = 1;
 	dma_dom->next_bit = 0;
 
+	dma_dom->need_flush = false;
+	dma_dom->target_dev = 0xffff;
+
 	/* Intialize the exclusion range if necessary */
 	if (iommu->exclusion_start &&
 	    iommu->exclusion_start < dma_dom->aperture_size) {
@@ -623,12 +745,13 @@ static void set_device_domain(struct amd_iommu *iommu,
 
 	u64 pte_root = virt_to_phys(domain->pt_root);
 
-	pte_root |= (domain->mode & 0x07) << 9;
-	pte_root |= IOMMU_PTE_IR | IOMMU_PTE_IW | IOMMU_PTE_P | 2;
+	pte_root |= (domain->mode & DEV_ENTRY_MODE_MASK)
+		    << DEV_ENTRY_MODE_SHIFT;
+	pte_root |= IOMMU_PTE_IR | IOMMU_PTE_IW | IOMMU_PTE_P | IOMMU_PTE_TV;
 
 	write_lock_irqsave(&amd_iommu_devtable_lock, flags);
-	amd_iommu_dev_table[devid].data[0] = pte_root;
-	amd_iommu_dev_table[devid].data[1] = pte_root >> 32;
+	amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root);
+	amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root);
 	amd_iommu_dev_table[devid].data[2] = domain->id;
 
 	amd_iommu_pd_table[devid] = domain;
@@ -646,6 +769,45 @@ static void set_device_domain(struct amd_iommu *iommu,
  *****************************************************************************/
 
 /*
+ * This function checks if the driver got a valid device from the caller to
+ * avoid dereferencing invalid pointers.
+ */
+static bool check_device(struct device *dev)
+{
+	if (!dev || !dev->dma_mask)
+		return false;
+
+	return true;
+}
+
+/*
+ * In this function the list of preallocated protection domains is traversed to
+ * find the domain for a specific device
+ */
+static struct dma_ops_domain *find_protection_domain(u16 devid)
+{
+	struct dma_ops_domain *entry, *ret = NULL;
+	unsigned long flags;
+
+	if (list_empty(&iommu_pd_list))
+		return NULL;
+
+	spin_lock_irqsave(&iommu_pd_list_lock, flags);
+
+	list_for_each_entry(entry, &iommu_pd_list, list) {
+		if (entry->target_dev == devid) {
+			ret = entry;
+			list_del(&ret->list);
+			break;
+		}
+	}
+
+	spin_unlock_irqrestore(&iommu_pd_list_lock, flags);
+
+	return ret;
+}
+
+/*
  * In the dma_ops path we only have the struct device. This function
  * finds the corresponding IOMMU, the protection domain and the
  * requestor id for a given device.
@@ -661,27 +823,30 @@ static int get_device_resources(struct device *dev,
 	struct pci_dev *pcidev;
 	u16 _bdf;
 
-	BUG_ON(!dev || dev->bus != &pci_bus_type || !dev->dma_mask);
+	*iommu = NULL;
+	*domain = NULL;
+	*bdf = 0xffff;
+
+	if (dev->bus != &pci_bus_type)
+		return 0;
 
 	pcidev = to_pci_dev(dev);
 	_bdf = calc_devid(pcidev->bus->number, pcidev->devfn);
 
 	/* device not translated by any IOMMU in the system? */
-	if (_bdf > amd_iommu_last_bdf) {
-		*iommu = NULL;
-		*domain = NULL;
-		*bdf = 0xffff;
+	if (_bdf > amd_iommu_last_bdf)
 		return 0;
-	}
 
 	*bdf = amd_iommu_alias_table[_bdf];
 
 	*iommu = amd_iommu_rlookup_table[*bdf];
 	if (*iommu == NULL)
 		return 0;
-	dma_dom = (*iommu)->default_dom;
 	*domain = domain_for_device(*bdf);
 	if (*domain == NULL) {
+		dma_dom = find_protection_domain(*bdf);
+		if (!dma_dom)
+			dma_dom = (*iommu)->default_dom;
 		*domain = &dma_dom->domain;
 		set_device_domain(*iommu, *domain, *bdf);
 		printk(KERN_INFO "AMD IOMMU: Using protection domain %d for "
@@ -760,17 +925,24 @@ static dma_addr_t __map_single(struct device *dev,
 			       struct dma_ops_domain *dma_dom,
 			       phys_addr_t paddr,
 			       size_t size,
-			       int dir)
+			       int dir,
+			       bool align,
+			       u64 dma_mask)
 {
 	dma_addr_t offset = paddr & ~PAGE_MASK;
 	dma_addr_t address, start;
 	unsigned int pages;
+	unsigned long align_mask = 0;
 	int i;
 
 	pages = iommu_num_pages(paddr, size);
 	paddr &= PAGE_MASK;
 
-	address = dma_ops_alloc_addresses(dev, dma_dom, pages);
+	if (align)
+		align_mask = (1UL << get_order(size)) - 1;
+
+	address = dma_ops_alloc_addresses(dev, dma_dom, pages, align_mask,
+					  dma_mask);
 	if (unlikely(address == bad_dma_address))
 		goto out;
 
@@ -782,6 +954,12 @@ static dma_addr_t __map_single(struct device *dev,
 	}
 	address += offset;
 
+	if (unlikely(dma_dom->need_flush && !amd_iommu_unmap_flush)) {
+		iommu_flush_tlb(iommu, dma_dom->domain.id);
+		dma_dom->need_flush = false;
+	} else if (unlikely(iommu_has_npcache(iommu)))
+		iommu_flush_pages(iommu, dma_dom->domain.id, address, size);
+
 out:
 	return address;
 }
@@ -812,6 +990,9 @@ static void __unmap_single(struct amd_iommu *iommu,
 	}
 
 	dma_ops_free_addresses(dma_dom, dma_addr, pages);
+
+	if (amd_iommu_unmap_flush)
+		iommu_flush_pages(iommu, dma_dom->domain.id, dma_addr, size);
 }
 
 /*
@@ -825,6 +1006,12 @@ static dma_addr_t map_single(struct device *dev, phys_addr_t paddr,
 	struct protection_domain *domain;
 	u16 devid;
 	dma_addr_t addr;
+	u64 dma_mask;
+
+	if (!check_device(dev))
+		return bad_dma_address;
+
+	dma_mask = *dev->dma_mask;
 
 	get_device_resources(dev, &iommu, &domain, &devid);
 
@@ -833,14 +1020,12 @@ static dma_addr_t map_single(struct device *dev, phys_addr_t paddr,
 		return (dma_addr_t)paddr;
 
 	spin_lock_irqsave(&domain->lock, flags);
-	addr = __map_single(dev, iommu, domain->priv, paddr, size, dir);
+	addr = __map_single(dev, iommu, domain->priv, paddr, size, dir, false,
+			    dma_mask);
 	if (addr == bad_dma_address)
 		goto out;
 
-	if (iommu_has_npcache(iommu))
-		iommu_flush_pages(iommu, domain->id, addr, size);
-
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 out:
@@ -860,7 +1045,8 @@ static void unmap_single(struct device *dev, dma_addr_t dma_addr,
 	struct protection_domain *domain;
 	u16 devid;
 
-	if (!get_device_resources(dev, &iommu, &domain, &devid))
+	if (!check_device(dev) ||
+	    !get_device_resources(dev, &iommu, &domain, &devid))
 		/* device not handled by any AMD IOMMU */
 		return;
 
@@ -868,9 +1054,7 @@ static void unmap_single(struct device *dev, dma_addr_t dma_addr,
 
 	__unmap_single(iommu, domain->priv, dma_addr, size, dir);
 
-	iommu_flush_pages(iommu, domain->id, dma_addr, size);
-
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 	spin_unlock_irqrestore(&domain->lock, flags);
@@ -909,6 +1093,12 @@ static int map_sg(struct device *dev, struct scatterlist *sglist,
 	struct scatterlist *s;
 	phys_addr_t paddr;
 	int mapped_elems = 0;
+	u64 dma_mask;
+
+	if (!check_device(dev))
+		return 0;
+
+	dma_mask = *dev->dma_mask;
 
 	get_device_resources(dev, &iommu, &domain, &devid);
 
@@ -921,19 +1111,17 @@ static int map_sg(struct device *dev, struct scatterlist *sglist,
 		paddr = sg_phys(s);
 
 		s->dma_address = __map_single(dev, iommu, domain->priv,
-					      paddr, s->length, dir);
+					      paddr, s->length, dir, false,
+					      dma_mask);
 
 		if (s->dma_address) {
 			s->dma_length = s->length;
 			mapped_elems++;
 		} else
 			goto unmap;
-		if (iommu_has_npcache(iommu))
-			iommu_flush_pages(iommu, domain->id, s->dma_address,
-					  s->dma_length);
 	}
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 out:
@@ -967,7 +1155,8 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist,
 	u16 devid;
 	int i;
 
-	if (!get_device_resources(dev, &iommu, &domain, &devid))
+	if (!check_device(dev) ||
+	    !get_device_resources(dev, &iommu, &domain, &devid))
 		return;
 
 	spin_lock_irqsave(&domain->lock, flags);
@@ -975,12 +1164,10 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist,
 	for_each_sg(sglist, s, nelems, i) {
 		__unmap_single(iommu, domain->priv, s->dma_address,
 			       s->dma_length, dir);
-		iommu_flush_pages(iommu, domain->id, s->dma_address,
-				  s->dma_length);
 		s->dma_address = s->dma_length = 0;
 	}
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 	spin_unlock_irqrestore(&domain->lock, flags);
@@ -998,25 +1185,33 @@ static void *alloc_coherent(struct device *dev, size_t size,
 	struct protection_domain *domain;
 	u16 devid;
 	phys_addr_t paddr;
+	u64 dma_mask = dev->coherent_dma_mask;
+
+	if (!check_device(dev))
+		return NULL;
 
+	if (!get_device_resources(dev, &iommu, &domain, &devid))
+		flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
+
+	flag |= __GFP_ZERO;
 	virt_addr = (void *)__get_free_pages(flag, get_order(size));
 	if (!virt_addr)
 		return 0;
 
-	memset(virt_addr, 0, size);
 	paddr = virt_to_phys(virt_addr);
 
-	get_device_resources(dev, &iommu, &domain, &devid);
-
 	if (!iommu || !domain) {
 		*dma_addr = (dma_addr_t)paddr;
 		return virt_addr;
 	}
 
+	if (!dma_mask)
+		dma_mask = *dev->dma_mask;
+
 	spin_lock_irqsave(&domain->lock, flags);
 
 	*dma_addr = __map_single(dev, iommu, domain->priv, paddr,
-				 size, DMA_BIDIRECTIONAL);
+				 size, DMA_BIDIRECTIONAL, true, dma_mask);
 
 	if (*dma_addr == bad_dma_address) {
 		free_pages((unsigned long)virt_addr, get_order(size));
@@ -1024,10 +1219,7 @@ static void *alloc_coherent(struct device *dev, size_t size,
 		goto out;
 	}
 
-	if (iommu_has_npcache(iommu))
-		iommu_flush_pages(iommu, domain->id, *dma_addr, size);
-
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 out:
@@ -1038,8 +1230,6 @@ out:
 
 /*
  * The exported free_coherent function for dma_ops.
- * FIXME: fix the generic x86 DMA layer so that it actually calls that
- *        function.
  */
 static void free_coherent(struct device *dev, size_t size,
 			  void *virt_addr, dma_addr_t dma_addr)
@@ -1049,6 +1239,9 @@ static void free_coherent(struct device *dev, size_t size,
 	struct protection_domain *domain;
 	u16 devid;
 
+	if (!check_device(dev))
+		return;
+
 	get_device_resources(dev, &iommu, &domain, &devid);
 
 	if (!iommu || !domain)
@@ -1057,9 +1250,8 @@ static void free_coherent(struct device *dev, size_t size,
 	spin_lock_irqsave(&domain->lock, flags);
 
 	__unmap_single(iommu, domain->priv, dma_addr, size, DMA_BIDIRECTIONAL);
-	iommu_flush_pages(iommu, domain->id, dma_addr, size);
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 	spin_unlock_irqrestore(&domain->lock, flags);
@@ -1069,6 +1261,30 @@ free_mem:
 }
 
 /*
+ * This function is called by the DMA layer to find out if we can handle a
+ * particular device. It is part of the dma_ops.
+ */
+static int amd_iommu_dma_supported(struct device *dev, u64 mask)
+{
+	u16 bdf;
+	struct pci_dev *pcidev;
+
+	/* No device or no PCI device */
+	if (!dev || dev->bus != &pci_bus_type)
+		return 0;
+
+	pcidev = to_pci_dev(dev);
+
+	bdf = calc_devid(pcidev->bus->number, pcidev->devfn);
+
+	/* Out of our scope? */
+	if (bdf > amd_iommu_last_bdf)
+		return 0;
+
+	return 1;
+}
+
+/*
  * The function for pre-allocating protection domains.
  *
  * If the driver core informs the DMA layer if a driver grabs a device
@@ -1097,10 +1313,9 @@ void prealloc_protection_domains(void)
 		if (!dma_dom)
 			continue;
 		init_unity_mappings_for_device(dma_dom, devid);
-		set_device_domain(iommu, &dma_dom->domain, devid);
-		printk(KERN_INFO "AMD IOMMU: Allocated domain %d for device ",
-		       dma_dom->domain.id);
-		print_devid(devid, 1);
+		dma_dom->target_dev = devid;
+
+		list_add_tail(&dma_dom->list, &iommu_pd_list);
 	}
 }
 
@@ -1111,6 +1326,7 @@ static struct dma_mapping_ops amd_iommu_dma_ops = {
 	.unmap_single = unmap_single,
 	.map_sg = map_sg,
 	.unmap_sg = unmap_sg,
+	.dma_supported = amd_iommu_dma_supported,
 };
 
 /*
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index a69cc0f5204..148fcfe22f1 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -22,6 +22,8 @@
 #include <linux/gfp.h>
 #include <linux/list.h>
 #include <linux/sysdev.h>
+#include <linux/interrupt.h>
+#include <linux/msi.h>
 #include <asm/pci-direct.h>
 #include <asm/amd_iommu_types.h>
 #include <asm/amd_iommu.h>
@@ -30,7 +32,6 @@
 /*
  * definitions for the ACPI scanning code
  */
-#define PCI_BUS(x) (((x) >> 8) & 0xff)
 #define IVRS_HEADER_LENGTH 48
 
 #define ACPI_IVHD_TYPE                  0x10
@@ -121,6 +122,7 @@ LIST_HEAD(amd_iommu_unity_map);		/* a list of required unity mappings
 					   we find in ACPI */
 unsigned amd_iommu_aperture_order = 26; /* size of aperture in power of 2 */
 int amd_iommu_isolate;			/* if 1, device isolation is enabled */
+bool amd_iommu_unmap_flush;		/* if true, flush on every unmap */
 
 LIST_HEAD(amd_iommu_list);		/* list of all AMD IOMMUs in the
 					   system */
@@ -234,7 +236,7 @@ static void __init iommu_feature_disable(struct amd_iommu *iommu, u8 bit)
 {
 	u32 ctrl;
 
-	ctrl = (u64)readl(iommu->mmio_base + MMIO_CONTROL_OFFSET);
+	ctrl = readl(iommu->mmio_base + MMIO_CONTROL_OFFSET);
 	ctrl &= ~(1 << bit);
 	writel(ctrl, iommu->mmio_base + MMIO_CONTROL_OFFSET);
 }
@@ -242,13 +244,23 @@ static void __init iommu_feature_disable(struct amd_iommu *iommu, u8 bit)
 /* Function to enable the hardware */
 void __init iommu_enable(struct amd_iommu *iommu)
 {
-	printk(KERN_INFO "AMD IOMMU: Enabling IOMMU at ");
-	print_devid(iommu->devid, 0);
-	printk(" cap 0x%hx\n", iommu->cap_ptr);
+	printk(KERN_INFO "AMD IOMMU: Enabling IOMMU "
+	       "at %02x:%02x.%x cap 0x%hx\n",
+	       iommu->dev->bus->number,
+	       PCI_SLOT(iommu->dev->devfn),
+	       PCI_FUNC(iommu->dev->devfn),
+	       iommu->cap_ptr);
 
 	iommu_feature_enable(iommu, CONTROL_IOMMU_EN);
 }
 
+/* Function to enable IOMMU event logging and event interrupts */
+void __init iommu_enable_event_logging(struct amd_iommu *iommu)
+{
+	iommu_feature_enable(iommu, CONTROL_EVT_LOG_EN);
+	iommu_feature_enable(iommu, CONTROL_EVT_INT_EN);
+}
+
 /*
  * mapping and unmapping functions for the IOMMU MMIO space. Each AMD IOMMU in
  * the system has one.
@@ -286,6 +298,14 @@ static void __init iommu_unmap_mmio_space(struct amd_iommu *iommu)
  ****************************************************************************/
 
 /*
+ * This function calculates the length of a given IVHD entry
+ */
+static inline int ivhd_entry_length(u8 *ivhd)
+{
+	return 0x04 << (*ivhd >> 6);
+}
+
+/*
  * This function reads the last device id the IOMMU has to handle from the PCI
  * capability header for this IOMMU
  */
@@ -329,7 +349,7 @@ static int __init find_last_devid_from_ivhd(struct ivhd_header *h)
 		default:
 			break;
 		}
-		p += 0x04 << (*p >> 6);
+		p += ivhd_entry_length(p);
 	}
 
 	WARN_ON(p != end);
@@ -414,7 +434,32 @@ static u8 * __init alloc_command_buffer(struct amd_iommu *iommu)
 
 static void __init free_command_buffer(struct amd_iommu *iommu)
 {
-	free_pages((unsigned long)iommu->cmd_buf, get_order(CMD_BUFFER_SIZE));
+	free_pages((unsigned long)iommu->cmd_buf,
+		   get_order(iommu->cmd_buf_size));
+}
+
+/* allocates the memory where the IOMMU will log its events to */
+static u8 * __init alloc_event_buffer(struct amd_iommu *iommu)
+{
+	u64 entry;
+	iommu->evt_buf = (u8 *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
+						get_order(EVT_BUFFER_SIZE));
+
+	if (iommu->evt_buf == NULL)
+		return NULL;
+
+	entry = (u64)virt_to_phys(iommu->evt_buf) | EVT_LEN_MASK;
+	memcpy_toio(iommu->mmio_base + MMIO_EVT_BUF_OFFSET,
+		    &entry, sizeof(entry));
+
+	iommu->evt_buf_size = EVT_BUFFER_SIZE;
+
+	return iommu->evt_buf;
+}
+
+static void __init free_event_buffer(struct amd_iommu *iommu)
+{
+	free_pages((unsigned long)iommu->evt_buf, get_order(EVT_BUFFER_SIZE));
 }
 
 /* sets a specific bit in the device table entry. */
@@ -487,19 +532,21 @@ static void __init set_device_exclusion_range(u16 devid, struct ivmd_header *m)
  */
 static void __init init_iommu_from_pci(struct amd_iommu *iommu)
 {
-	int bus = PCI_BUS(iommu->devid);
-	int dev = PCI_SLOT(iommu->devid);
-	int fn  = PCI_FUNC(iommu->devid);
 	int cap_ptr = iommu->cap_ptr;
-	u32 range;
+	u32 range, misc;
 
-	iommu->cap = read_pci_config(bus, dev, fn, cap_ptr+MMIO_CAP_HDR_OFFSET);
+	pci_read_config_dword(iommu->dev, cap_ptr + MMIO_CAP_HDR_OFFSET,
+			      &iommu->cap);
+	pci_read_config_dword(iommu->dev, cap_ptr + MMIO_RANGE_OFFSET,
+			      &range);
+	pci_read_config_dword(iommu->dev, cap_ptr + MMIO_MISC_OFFSET,
+			      &misc);
 
-	range = read_pci_config(bus, dev, fn, cap_ptr+MMIO_RANGE_OFFSET);
 	iommu->first_device = calc_devid(MMIO_GET_BUS(range),
 					 MMIO_GET_FD(range));
 	iommu->last_device = calc_devid(MMIO_GET_BUS(range),
 					MMIO_GET_LD(range));
+	iommu->evt_msi_num = MMIO_MSI_NUM(misc);
 }
 
 /*
@@ -604,7 +651,7 @@ static void __init init_iommu_from_acpi(struct amd_iommu *iommu,
 			break;
 		}
 
-		p += 0x04 << (e->type >> 6);
+		p += ivhd_entry_length(p);
 	}
 }
 
@@ -622,6 +669,7 @@ static int __init init_iommu_devices(struct amd_iommu *iommu)
 static void __init free_iommu_one(struct amd_iommu *iommu)
 {
 	free_command_buffer(iommu);
+	free_event_buffer(iommu);
 	iommu_unmap_mmio_space(iommu);
 }
 
@@ -649,8 +697,12 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h)
 	/*
 	 * Copy data from ACPI table entry to the iommu struct
 	 */
-	iommu->devid = h->devid;
+	iommu->dev = pci_get_bus_and_slot(PCI_BUS(h->devid), h->devid & 0xff);
+	if (!iommu->dev)
+		return 1;
+
 	iommu->cap_ptr = h->cap_ptr;
+	iommu->pci_seg = h->pci_seg;
 	iommu->mmio_phys = h->mmio_phys;
 	iommu->mmio_base = iommu_map_mmio_space(h->mmio_phys);
 	if (!iommu->mmio_base)
@@ -661,10 +713,18 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h)
 	if (!iommu->cmd_buf)
 		return -ENOMEM;
 
+	iommu->evt_buf = alloc_event_buffer(iommu);
+	if (!iommu->evt_buf)
+		return -ENOMEM;
+
+	iommu->int_enabled = false;
+
 	init_iommu_from_pci(iommu);
 	init_iommu_from_acpi(iommu, h);
 	init_iommu_devices(iommu);
 
+	pci_enable_device(iommu->dev);
+
 	return 0;
 }
 
@@ -706,6 +766,95 @@ static int __init init_iommu_all(struct acpi_table_header *table)
 
 /****************************************************************************
  *
+ * The following functions initialize the MSI interrupts for all IOMMUs
+ * in the system. Its a bit challenging because there could be multiple
+ * IOMMUs per PCI BDF but we can call pci_enable_msi(x) only once per
+ * pci_dev.
+ *
+ ****************************************************************************/
+
+static int __init iommu_setup_msix(struct amd_iommu *iommu)
+{
+	struct amd_iommu *curr;
+	struct msix_entry entries[32]; /* only 32 supported by AMD IOMMU */
+	int nvec = 0, i;
+
+	list_for_each_entry(curr, &amd_iommu_list, list) {
+		if (curr->dev == iommu->dev) {
+			entries[nvec].entry = curr->evt_msi_num;
+			entries[nvec].vector = 0;
+			curr->int_enabled = true;
+			nvec++;
+		}
+	}
+
+	if (pci_enable_msix(iommu->dev, entries, nvec)) {
+		pci_disable_msix(iommu->dev);
+		return 1;
+	}
+
+	for (i = 0; i < nvec; ++i) {
+		int r = request_irq(entries->vector, amd_iommu_int_handler,
+				    IRQF_SAMPLE_RANDOM,
+				    "AMD IOMMU",
+				    NULL);
+		if (r)
+			goto out_free;
+	}
+
+	return 0;
+
+out_free:
+	for (i -= 1; i >= 0; --i)
+		free_irq(entries->vector, NULL);
+
+	pci_disable_msix(iommu->dev);
+
+	return 1;
+}
+
+static int __init iommu_setup_msi(struct amd_iommu *iommu)
+{
+	int r;
+	struct amd_iommu *curr;
+
+	list_for_each_entry(curr, &amd_iommu_list, list) {
+		if (curr->dev == iommu->dev)
+			curr->int_enabled = true;
+	}
+
+
+	if (pci_enable_msi(iommu->dev))
+		return 1;
+
+	r = request_irq(iommu->dev->irq, amd_iommu_int_handler,
+			IRQF_SAMPLE_RANDOM,
+			"AMD IOMMU",
+			NULL);
+
+	if (r) {
+		pci_disable_msi(iommu->dev);
+		return 1;
+	}
+
+	return 0;
+}
+
+static int __init iommu_init_msi(struct amd_iommu *iommu)
+{
+	if (iommu->int_enabled)
+		return 0;
+
+	if (pci_find_capability(iommu->dev, PCI_CAP_ID_MSIX))
+		return iommu_setup_msix(iommu);
+	else if (pci_find_capability(iommu->dev, PCI_CAP_ID_MSI))
+		return iommu_setup_msi(iommu);
+
+	return 1;
+}
+
+/****************************************************************************
+ *
  * The next functions belong to the third pass of parsing the ACPI
  * table. In this last pass the memory mapping requirements are
  * gathered (like exclusion and unity mapping reanges).
@@ -811,7 +960,6 @@ static void init_device_table(void)
 	for (devid = 0; devid <= amd_iommu_last_bdf; ++devid) {
 		set_dev_entry_bit(devid, DEV_ENTRY_VALID);
 		set_dev_entry_bit(devid, DEV_ENTRY_TRANSLATION);
-		set_dev_entry_bit(devid, DEV_ENTRY_NO_PAGE_FAULT);
 	}
 }
 
@@ -825,6 +973,8 @@ static void __init enable_iommus(void)
 
 	list_for_each_entry(iommu, &amd_iommu_list, list) {
 		iommu_set_exclusion_range(iommu);
+		iommu_init_msi(iommu);
+		iommu_enable_event_logging(iommu);
 		iommu_enable(iommu);
 	}
 }
@@ -995,11 +1145,17 @@ int __init amd_iommu_init(void)
 	else
 		printk("disabled\n");
 
+	if (amd_iommu_unmap_flush)
+		printk(KERN_INFO "AMD IOMMU: IO/TLB flush on unmap enabled\n");
+	else
+		printk(KERN_INFO "AMD IOMMU: Lazy IO/TLB flushing enabled\n");
+
 out:
 	return ret;
 
 free:
-	free_pages((unsigned long)amd_iommu_pd_alloc_bitmap, 1);
+	free_pages((unsigned long)amd_iommu_pd_alloc_bitmap,
+		   get_order(MAX_DOMAIN_ID/8));
 
 	free_pages((unsigned long)amd_iommu_pd_table,
 		   get_order(rlookup_table_size));
@@ -1057,8 +1213,10 @@ void __init amd_iommu_detect(void)
 static int __init parse_amd_iommu_options(char *str)
 {
 	for (; *str; ++str) {
-		if (strcmp(str, "isolate") == 0)
+		if (strncmp(str, "isolate", 7) == 0)
 			amd_iommu_isolate = 1;
+		if (strncmp(str, "fullflush", 11) == 0)
+			amd_iommu_unmap_flush = true;
 	}
 
 	return 1;
diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c
index 44e21826db1..9a32b37ee2e 100644
--- a/arch/x86/kernel/aperture_64.c
+++ b/arch/x86/kernel/aperture_64.c
@@ -455,11 +455,11 @@ out:
 		   force_iommu ||
 		   valid_agp ||
 		   fallback_aper_force) {
-		printk(KERN_ERR
+		printk(KERN_INFO
 			"Your BIOS doesn't leave a aperture memory hole\n");
-		printk(KERN_ERR
+		printk(KERN_INFO
 			"Please enable the IOMMU option in the BIOS setup\n");
-		printk(KERN_ERR
+		printk(KERN_INFO
 			"This costs you %d MB of RAM\n",
 				32 << fallback_aper_order);
 
diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index 58427210505..a91c57cb666 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -60,10 +60,8 @@ unsigned long mp_lapic_addr;
 static int force_enable_local_apic;
 int disable_apic;
 
-/* Local APIC timer verification ok */
-static int local_apic_timer_verify_ok;
 /* Disable local APIC timer from the kernel commandline or via dmi quirk */
-static int local_apic_timer_disabled;
+static int disable_apic_timer __cpuinitdata;
 /* Local APIC timer works in C2 */
 int local_apic_timer_c2_ok;
 EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok);
@@ -130,7 +128,11 @@ static inline int lapic_get_version(void)
  */
 static inline int lapic_is_integrated(void)
 {
+#ifdef CONFIG_X86_64
+	return 1;
+#else
 	return APIC_INTEGRATED(lapic_get_version());
+#endif
 }
 
 /*
@@ -244,8 +246,12 @@ int lapic_get_maxlvt(void)
  * Local APIC timer
  */
 
-/* Clock divisor is set to 16 */
+/* Clock divisor */
+#ifdef CONFG_X86_64
+#define APIC_DIVISOR 1
+#else
 #define APIC_DIVISOR 16
+#endif
 
 /*
  * This function sets up the local APIC timer, with a timeout of
@@ -253,6 +259,9 @@ int lapic_get_maxlvt(void)
  * this function twice on the boot CPU, once with a bogus timeout
  * value, second time for real. The other (noncalibrating) CPUs
  * call this function only once, with the real, calibrated value.
+ *
+ * We do reads before writes even if unnecessary, to get around the
+ * P5 APIC double write bug.
  */
 static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 {
@@ -274,14 +283,44 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 	 */
 	tmp_value = apic_read(APIC_TDCR);
 	apic_write(APIC_TDCR,
-		   (tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) |
-		   APIC_TDR_DIV_16);
+		(tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) |
+		APIC_TDR_DIV_16);
 
 	if (!oneshot)
 		apic_write(APIC_TMICT, clocks / APIC_DIVISOR);
 }
 
 /*
+ * Setup extended LVT, AMD specific (K8, family 10h)
+ *
+ * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and
+ * MCE interrupts are supported. Thus MCE offset must be set to 0.
+ */
+
+#define APIC_EILVT_LVTOFF_MCE 0
+#define APIC_EILVT_LVTOFF_IBS 1
+
+static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask)
+{
+	unsigned long reg = (lvt_off << 4) + APIC_EILVT0;
+	unsigned int  v   = (mask << 16) | (msg_type << 8) | vector;
+
+	apic_write(reg, v);
+}
+
+u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask)
+{
+	setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask);
+	return APIC_EILVT_LVTOFF_MCE;
+}
+
+u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask)
+{
+	setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask);
+	return APIC_EILVT_LVTOFF_IBS;
+}
+
+/*
  * Program the next event, relative to now
  */
 static int lapic_next_event(unsigned long delta,
@@ -300,8 +339,8 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 	unsigned long flags;
 	unsigned int v;
 
-	/* Lapic used for broadcast ? */
-	if (!local_apic_timer_verify_ok)
+	/* Lapic used as dummy for broadcast ? */
+	if (evt->features & CLOCK_EVT_FEAT_DUMMY)
 		return;
 
 	local_irq_save(flags);
@@ -514,7 +553,7 @@ static int __init calibrate_APIC_clock(void)
 		return -1;
 	}
 
-	local_apic_timer_verify_ok = 1;
+	levt->features &= ~CLOCK_EVT_FEAT_DUMMY;
 
 	/* We trust the pm timer based calibration */
 	if (!pm_referenced) {
@@ -548,11 +587,11 @@ static int __init calibrate_APIC_clock(void)
 		if (deltaj >= LAPIC_CAL_LOOPS-2 && deltaj <= LAPIC_CAL_LOOPS+2)
 			apic_printk(APIC_VERBOSE, "... jiffies result ok\n");
 		else
-			local_apic_timer_verify_ok = 0;
+			levt->features |= CLOCK_EVT_FEAT_DUMMY;
 	} else
 		local_irq_enable();
 
-	if (!local_apic_timer_verify_ok) {
+	if (levt->features & CLOCK_EVT_FEAT_DUMMY) {
 		printk(KERN_WARNING
 		       "APIC timer disabled due to verification failure.\n");
 			return -1;
@@ -574,7 +613,8 @@ void __init setup_boot_APIC_clock(void)
 	 * timer as a dummy clock event source on SMP systems, so the
 	 * broadcast mechanism is used. On UP systems simply ignore it.
 	 */
-	if (local_apic_timer_disabled) {
+	if (disable_apic_timer) {
+		printk(KERN_INFO "Disabling APIC timer\n");
 		/* No broadcast on UP ! */
 		if (num_possible_cpus() > 1) {
 			lapic_clockevent.mult = 1;
@@ -643,7 +683,11 @@ static void local_apic_timer_interrupt(void)
 	/*
 	 * the NMI deadlock-detector uses this.
 	 */
+#ifdef CONFIG_X86_64
+	add_pda(apic_timer_irqs, 1);
+#else
 	per_cpu(irq_stat, cpu).apic_timer_irqs++;
+#endif
 
 	evt->event_handler(evt);
 }
@@ -683,35 +727,6 @@ int setup_profiling_timer(unsigned int multiplier)
 }
 
 /*
- * Setup extended LVT, AMD specific (K8, family 10h)
- *
- * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and
- * MCE interrupts are supported. Thus MCE offset must be set to 0.
- */
-
-#define APIC_EILVT_LVTOFF_MCE 0
-#define APIC_EILVT_LVTOFF_IBS 1
-
-static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask)
-{
-	unsigned long reg = (lvt_off << 4) + APIC_EILVT0;
-	unsigned int  v   = (mask << 16) | (msg_type << 8) | vector;
-	apic_write(reg, v);
-}
-
-u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask)
-{
-	setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask);
-	return APIC_EILVT_LVTOFF_MCE;
-}
-
-u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask)
-{
-	setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask);
-	return APIC_EILVT_LVTOFF_IBS;
-}
-
-/*
  * Local APIC start and shutdown
  */
 
@@ -756,7 +771,7 @@ void clear_local_APIC(void)
 	}
 
 	/* lets not touch this if we didn't frob it */
-#ifdef CONFIG_X86_MCE_P4THERMAL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL)
 	if (maxlvt >= 5) {
 		v = apic_read(APIC_LVTTHMR);
 		apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED);
@@ -773,10 +788,6 @@ void clear_local_APIC(void)
 	if (maxlvt >= 4)
 		apic_write(APIC_LVTPC, APIC_LVT_MASKED);
 
-#ifdef CONFIG_X86_MCE_P4THERMAL
-	if (maxlvt >= 5)
-		apic_write(APIC_LVTTHMR, APIC_LVT_MASKED);
-#endif
 	/* Integrated APIC (!82489DX) ? */
 	if (lapic_is_integrated()) {
 		if (maxlvt > 3)
@@ -791,7 +802,7 @@ void clear_local_APIC(void)
  */
 void disable_local_APIC(void)
 {
-	unsigned long value;
+	unsigned int value;
 
 	clear_local_APIC();
 
@@ -803,6 +814,7 @@ void disable_local_APIC(void)
 	value &= ~APIC_SPIV_APIC_ENABLED;
 	apic_write(APIC_SPIV, value);
 
+#ifdef CONFIG_X86_32
 	/*
 	 * When LAPIC was disabled by the BIOS and enabled by the kernel,
 	 * restore the disabled state.
@@ -814,6 +826,7 @@ void disable_local_APIC(void)
 		l &= ~MSR_IA32_APICBASE_ENABLE;
 		wrmsr(MSR_IA32_APICBASE, l, h);
 	}
+#endif
 }
 
 /*
@@ -830,11 +843,15 @@ void lapic_shutdown(void)
 		return;
 
 	local_irq_save(flags);
-	clear_local_APIC();
 
-	if (enabled_via_apicbase)
+#ifdef CONFIG_X86_32
+	if (!enabled_via_apicbase)
+		clear_local_APIC();
+	else
+#endif
 		disable_local_APIC();
 
+
 	local_irq_restore(flags);
 }
 
@@ -879,6 +896,12 @@ int __init verify_local_APIC(void)
 	 */
 	reg0 = apic_read(APIC_ID);
 	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0);
+	apic_write(APIC_ID, reg0 ^ APIC_ID_MASK);
+	reg1 = apic_read(APIC_ID);
+	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1);
+	apic_write(APIC_ID, reg0);
+	if (reg1 != (reg0 ^ APIC_ID_MASK))
+		return 0;
 
 	/*
 	 * The next two are just to see if we have sane values.
@@ -904,14 +927,15 @@ void __init sync_Arb_IDs(void)
 	 */
 	if (modern_apic() || boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
 		return;
+
 	/*
 	 * Wait for idle.
 	 */
 	apic_wait_icr_idle();
 
 	apic_printk(APIC_DEBUG, "Synchronizing Arb IDs.\n");
-	apic_write(APIC_ICR,
-		   APIC_DEST_ALLINC | APIC_INT_LEVELTRIG | APIC_DM_INIT);
+	apic_write(APIC_ICR, APIC_DEST_ALLINC |
+			APIC_INT_LEVELTRIG | APIC_DM_INIT);
 }
 
 /*
@@ -919,7 +943,7 @@ void __init sync_Arb_IDs(void)
  */
 void __init init_bsp_APIC(void)
 {
-	unsigned long value;
+	unsigned int value;
 
 	/*
 	 * Don't do the setup now if we have a SMP BIOS as the
@@ -940,11 +964,13 @@ void __init init_bsp_APIC(void)
 	value &= ~APIC_VECTOR_MASK;
 	value |= APIC_SPIV_APIC_ENABLED;
 
+#ifdef CONFIG_X86_32
 	/* This bit is reserved on P4/Xeon and should be cleared */
 	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
 	    (boot_cpu_data.x86 == 15))
 		value &= ~APIC_SPIV_FOCUS_DISABLED;
 	else
+#endif
 		value |= APIC_SPIV_FOCUS_DISABLED;
 	value |= SPURIOUS_APIC_VECTOR;
 	apic_write(APIC_SPIV, value);
@@ -963,6 +989,16 @@ static void __cpuinit lapic_setup_esr(void)
 {
 	unsigned long oldvalue, value, maxlvt;
 	if (lapic_is_integrated() && !esr_disable) {
+		if (esr_disable) {
+			/*
+			 * Something untraceable is creating bad interrupts on
+			 * secondary quads ... for the moment, just leave the
+			 * ESR disabled - we can't do anything useful with the
+			 * errors anyway - mbligh
+			 */
+			printk(KERN_INFO "Leaving ESR disabled.\n");
+			return;
+		}
 		/* !82489DX */
 		maxlvt = lapic_get_maxlvt();
 		if (maxlvt > 3)		/* Due to the Pentium erratum 3AP. */
@@ -983,16 +1019,7 @@ static void __cpuinit lapic_setup_esr(void)
 				"vector: 0x%08lx  after: 0x%08lx\n",
 				oldvalue, value);
 	} else {
-		if (esr_disable)
-			/*
-			 * Something untraceable is creating bad interrupts on
-			 * secondary quads ... for the moment, just leave the
-			 * ESR disabled - we can't do anything useful with the
-			 * errors anyway - mbligh
-			 */
-			printk(KERN_INFO "Leaving ESR disabled.\n");
-		else
-			printk(KERN_INFO "No ESR for 82489DX.\n");
+		printk(KERN_INFO "No ESR for 82489DX.\n");
 	}
 }
 
@@ -1130,13 +1157,17 @@ void __cpuinit setup_local_APIC(void)
 
 void __cpuinit end_local_APIC_setup(void)
 {
-	unsigned long value;
-
 	lapic_setup_esr();
-	/* Disable the local apic timer */
-	value = apic_read(APIC_LVTT);
-	value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
-	apic_write(APIC_LVTT, value);
+
+#ifdef CONFIG_X86_32
+	{
+		unsigned int value;
+		/* Disable the local apic timer */
+		value = apic_read(APIC_LVTT);
+		value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
+		apic_write(APIC_LVTT, value);
+	}
+#endif
 
 	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
@@ -1367,6 +1398,7 @@ void smp_error_interrupt(struct pt_regs *regs)
  */
 void __init connect_bsp_APIC(void)
 {
+#ifdef CONFIG_X86_32
 	if (pic_mode) {
 		/*
 		 * Do not trust the local APIC being empty at bootup.
@@ -1381,6 +1413,7 @@ void __init connect_bsp_APIC(void)
 		outb(0x70, 0x22);
 		outb(0x01, 0x23);
 	}
+#endif
 	enable_apic_mode();
 }
 
@@ -1393,6 +1426,9 @@ void __init connect_bsp_APIC(void)
  */
 void disconnect_bsp_APIC(int virt_wire_setup)
 {
+	unsigned int value;
+
+#ifdef CONFIG_X86_32
 	if (pic_mode) {
 		/*
 		 * Put the board back into PIC mode (has an effect only on
@@ -1404,54 +1440,53 @@ void disconnect_bsp_APIC(int virt_wire_setup)
 				"entering PIC mode.\n");
 		outb(0x70, 0x22);
 		outb(0x00, 0x23);
-	} else {
-		/* Go back to Virtual Wire compatibility mode */
-		unsigned long value;
+		return;
+	}
+#endif
 
-		/* For the spurious interrupt use vector F, and enable it */
-		value = apic_read(APIC_SPIV);
-		value &= ~APIC_VECTOR_MASK;
-		value |= APIC_SPIV_APIC_ENABLED;
-		value |= 0xf;
-		apic_write(APIC_SPIV, value);
+	/* Go back to Virtual Wire compatibility mode */
 
-		if (!virt_wire_setup) {
-			/*
-			 * For LVT0 make it edge triggered, active high,
-			 * external and enabled
-			 */
-			value = apic_read(APIC_LVT0);
-			value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
-				APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
-				APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
-			value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-			value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
-			apic_write(APIC_LVT0, value);
-		} else {
-			/* Disable LVT0 */
-			apic_write(APIC_LVT0, APIC_LVT_MASKED);
-		}
+	/* For the spurious interrupt use vector F, and enable it */
+	value = apic_read(APIC_SPIV);
+	value &= ~APIC_VECTOR_MASK;
+	value |= APIC_SPIV_APIC_ENABLED;
+	value |= 0xf;
+	apic_write(APIC_SPIV, value);
 
+	if (!virt_wire_setup) {
 		/*
-		 * For LVT1 make it edge triggered, active high, nmi and
-		 * enabled
+		 * For LVT0 make it edge triggered, active high,
+		 * external and enabled
 		 */
-		value = apic_read(APIC_LVT1);
-		value &= ~(
-			APIC_MODE_MASK | APIC_SEND_PENDING |
+		value = apic_read(APIC_LVT0);
+		value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
 			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
 			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
 		value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
-		apic_write(APIC_LVT1, value);
+		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
+		apic_write(APIC_LVT0, value);
+	} else {
+		/* Disable LVT0 */
+		apic_write(APIC_LVT0, APIC_LVT_MASKED);
 	}
+
+	/*
+	 * For LVT1 make it edge triggered, active high,
+	 * nmi and enabled
+	 */
+	value = apic_read(APIC_LVT1);
+	value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
+			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+	value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+	value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
+	apic_write(APIC_LVT1, value);
 }
 
 void __cpuinit generic_processor_info(int apicid, int version)
 {
 	int cpu;
 	cpumask_t tmp_map;
-	physid_mask_t phys_cpu;
 
 	/*
 	 * Validate version
@@ -1464,9 +1499,6 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	}
 	apic_version[apicid] = version;
 
-	phys_cpu = apicid_to_cpu_present(apicid);
-	physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu);
-
 	if (num_processors >= NR_CPUS) {
 		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
 			"  Processor ignored.\n", NR_CPUS);
@@ -1477,17 +1509,19 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	cpus_complement(tmp_map, cpu_present_map);
 	cpu = first_cpu(tmp_map);
 
-	if (apicid == boot_cpu_physical_apicid)
+	physid_set(apicid, phys_cpu_present_map);
+	if (apicid == boot_cpu_physical_apicid) {
 		/*
 		 * x86_bios_cpu_apicid is required to have processors listed
 		 * in same order as logical cpu numbers. Hence the first
 		 * entry is BSP, and so on.
 		 */
 		cpu = 0;
-
+	}
 	if (apicid > max_physical_apicid)
 		max_physical_apicid = apicid;
 
+#ifdef CONFIG_X86_32
 	/*
 	 * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
 	 * but we need to work other dependencies like SMP_SUSPEND etc
@@ -1507,7 +1541,9 @@ void __cpuinit generic_processor_info(int apicid, int version)
 			def_to_bigsmp = 1;
 		}
 	}
-#ifdef CONFIG_SMP
+#endif
+
+#if defined(CONFIG_X86_SMP) || defined(CONFIG_X86_64)
 	/* are we being called early in kernel startup? */
 	if (early_per_cpu_ptr(x86_cpu_to_apicid)) {
 		u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);
@@ -1520,6 +1556,7 @@ void __cpuinit generic_processor_info(int apicid, int version)
 		per_cpu(x86_bios_cpu_apicid, cpu) = apicid;
 	}
 #endif
+
 	cpu_set(cpu, cpu_possible_map);
 	cpu_set(cpu, cpu_present_map);
 }
@@ -1530,6 +1567,11 @@ void __cpuinit generic_processor_info(int apicid, int version)
 #ifdef CONFIG_PM
 
 static struct {
+	/*
+	 * 'active' is true if the local APIC was enabled by us and
+	 * not the BIOS; this signifies that we are also responsible
+	 * for disabling it before entering apm/acpi suspend
+	 */
 	int active;
 	/* r/w apic fields */
 	unsigned int apic_id;
@@ -1570,7 +1612,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 	apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR);
 	apic_pm_state.apic_tmict = apic_read(APIC_TMICT);
 	apic_pm_state.apic_tdcr = apic_read(APIC_TDCR);
-#ifdef CONFIG_X86_MCE_P4THERMAL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR);
 #endif
@@ -1594,16 +1636,23 @@ static int lapic_resume(struct sys_device *dev)
 
 	local_irq_save(flags);
 
-	/*
-	 * Make sure the APICBASE points to the right address
-	 *
-	 * FIXME! This will be wrong if we ever support suspend on
-	 * SMP! We'll need to do this as part of the CPU restore!
-	 */
-	rdmsr(MSR_IA32_APICBASE, l, h);
-	l &= ~MSR_IA32_APICBASE_BASE;
-	l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr;
-	wrmsr(MSR_IA32_APICBASE, l, h);
+#ifdef CONFIG_X86_64
+	if (x2apic)
+		enable_x2apic();
+	else
+#endif
+	{
+		/*
+		 * Make sure the APICBASE points to the right address
+		 *
+		 * FIXME! This will be wrong if we ever support suspend on
+		 * SMP! We'll need to do this as part of the CPU restore!
+		 */
+		rdmsr(MSR_IA32_APICBASE, l, h);
+		l &= ~MSR_IA32_APICBASE_BASE;
+		l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr;
+		wrmsr(MSR_IA32_APICBASE, l, h);
+	}
 
 	apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED);
 	apic_write(APIC_ID, apic_pm_state.apic_id);
@@ -1613,7 +1662,7 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_SPIV, apic_pm_state.apic_spiv);
 	apic_write(APIC_LVT0, apic_pm_state.apic_lvt0);
 	apic_write(APIC_LVT1, apic_pm_state.apic_lvt1);
-#ifdef CONFIG_X86_MCE_P4THERMAL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr);
 #endif
@@ -1627,7 +1676,9 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_LVTERR, apic_pm_state.apic_lvterr);
 	apic_write(APIC_ESR, 0);
 	apic_read(APIC_ESR);
+
 	local_irq_restore(flags);
+
 	return 0;
 }
 
@@ -1683,20 +1734,20 @@ static int __init parse_lapic(char *arg)
 }
 early_param("lapic", parse_lapic);
 
-static int __init parse_nolapic(char *arg)
+static int __init setup_disableapic(char *arg)
 {
 	disable_apic = 1;
 	setup_clear_cpu_cap(X86_FEATURE_APIC);
 	return 0;
 }
-early_param("nolapic", parse_nolapic);
+early_param("disableapic", setup_disableapic);
 
-static int __init parse_disable_lapic_timer(char *arg)
+/* same as disableapic, for compatibility */
+static int __init setup_nolapic(char *arg)
 {
-	local_apic_timer_disabled = 1;
-	return 0;
+	return setup_disableapic(arg);
 }
-early_param("nolapic_timer", parse_disable_lapic_timer);
+early_param("nolapic", setup_nolapic);
 
 static int __init parse_lapic_timer_c2_ok(char *arg)
 {
@@ -1705,15 +1756,40 @@ static int __init parse_lapic_timer_c2_ok(char *arg)
 }
 early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok);
 
+static int __init parse_disable_apic_timer(char *arg)
+{
+	disable_apic_timer = 1;
+	return 0;
+}
+early_param("noapictimer", parse_disable_apic_timer);
+
+static int __init parse_nolapic_timer(char *arg)
+{
+	disable_apic_timer = 1;
+	return 0;
+}
+early_param("nolapic_timer", parse_nolapic_timer);
+
 static int __init apic_set_verbosity(char *arg)
 {
-	if (!arg)
+	if (!arg)  {
+#ifdef CONFIG_X86_64
+		skip_ioapic_setup = 0;
+		ioapic_force = 1;
+		return 0;
+#endif
 		return -EINVAL;
+	}
 
-	if (strcmp(arg, "debug") == 0)
+	if (strcmp("debug", arg) == 0)
 		apic_verbosity = APIC_DEBUG;
-	else if (strcmp(arg, "verbose") == 0)
+	else if (strcmp("verbose", arg) == 0)
 		apic_verbosity = APIC_VERBOSE;
+	else {
+		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
+			" use apic=verbose or apic=debug\n", arg);
+		return -EINVAL;
+	}
 
 	return 0;
 }
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 1a6011855af..53898b65a6a 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -45,6 +45,7 @@
 #include <mach_ipi.h>
 #include <mach_apic.h>
 
+/* Disable local APIC timer from the kernel commandline or via dmi quirk */
 static int disable_apic_timer __cpuinitdata;
 static int apic_calibrate_pmtmr __initdata;
 int disable_apic;
@@ -80,6 +81,9 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 static void lapic_timer_broadcast(cpumask_t mask);
 static void apic_pm_activate(void);
 
+/*
+ * The local apic timer can be used for any function which is CPU local.
+ */
 static struct clock_event_device lapic_clockevent = {
 	.name		= "lapic",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT
@@ -106,11 +110,15 @@ static inline int lapic_get_version(void)
 }
 
 /*
- * Check, if the APIC is integrated or a seperate chip
+ * Check, if the APIC is integrated or a separate chip
  */
 static inline int lapic_is_integrated(void)
 {
+#ifdef CONFIG_X86_64
 	return 1;
+#else
+	return APIC_INTEGRATED(lapic_get_version());
+#endif
 }
 
 /*
@@ -125,6 +133,11 @@ static int modern_apic(void)
 	return lapic_get_version() >= 0x14;
 }
 
+/*
+ * Paravirt kernels also might be using these below ops. So we still
+ * use generic apic_read()/apic_write(), which might be pointing to different
+ * ops in PARAVIRT case.
+ */
 void xapic_wait_icr_idle(void)
 {
 	while (apic_read(APIC_ICR) & APIC_ICR_BUSY)
@@ -149,7 +162,7 @@ u32 safe_xapic_wait_icr_idle(void)
 
 void xapic_icr_write(u32 low, u32 id)
 {
-	apic_write(APIC_ICR2, id << 24);
+	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id));
 	apic_write(APIC_ICR, low);
 }
 
@@ -160,7 +173,7 @@ u64 xapic_icr_read(void)
 	icr2 = apic_read(APIC_ICR2);
 	icr1 = apic_read(APIC_ICR);
 
-	return (icr1 | ((u64)icr2 << 32));
+	return icr1 | ((u64)icr2 << 32);
 }
 
 static struct apic_ops xapic_ops = {
@@ -173,7 +186,6 @@ static struct apic_ops xapic_ops = {
 };
 
 struct apic_ops __read_mostly *apic_ops = &xapic_ops;
-
 EXPORT_SYMBOL_GPL(apic_ops);
 
 static void x2apic_wait_icr_idle(void)
@@ -243,6 +255,17 @@ int lapic_get_maxlvt(void)
 }
 
 /*
+ * Local APIC timer
+ */
+
+/* Clock divisor */
+#ifdef CONFG_X86_64
+#define APIC_DIVISOR 1
+#else
+#define APIC_DIVISOR 16
+#endif
+
+/*
  * This function sets up the local APIC timer, with a timeout of
  * 'clocks' APIC bus clock. During calibration we actually call
  * this function twice on the boot CPU, once with a bogus timeout
@@ -252,7 +275,6 @@ int lapic_get_maxlvt(void)
  * We do reads before writes even if unnecessary, to get around the
  * P5 APIC double write bug.
  */
-
 static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 {
 	unsigned int lvtt_value, tmp_value;
@@ -260,6 +282,9 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 	lvtt_value = LOCAL_TIMER_VECTOR;
 	if (!oneshot)
 		lvtt_value |= APIC_LVT_TIMER_PERIODIC;
+	if (!lapic_is_integrated())
+		lvtt_value |= SET_APIC_TIMER_BASE(APIC_TIMER_BASE_DIV);
+
 	if (!irqen)
 		lvtt_value |= APIC_LVT_MASKED;
 
@@ -269,12 +294,12 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 	 * Divide PICLK by 16
 	 */
 	tmp_value = apic_read(APIC_TDCR);
-	apic_write(APIC_TDCR, (tmp_value
-				& ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE))
-				| APIC_TDR_DIV_16);
+	apic_write(APIC_TDCR,
+		(tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) |
+		APIC_TDR_DIV_16);
 
 	if (!oneshot)
-		apic_write(APIC_TMICT, clocks);
+		apic_write(APIC_TMICT, clocks / APIC_DIVISOR);
 }
 
 /*
@@ -444,7 +469,7 @@ static int __init calibrate_APIC_clock(void)
 	lapic_clockevent.min_delta_ns =
 		clockevent_delta2ns(0xF, &lapic_clockevent);
 
-	calibration_result = result / HZ;
+	calibration_result = (result * APIC_DIVISOR) / HZ;
 
 	/*
 	 * Do a sanity check on the APIC calibration result
@@ -466,10 +491,10 @@ static int __init calibrate_APIC_clock(void)
 void __init setup_boot_APIC_clock(void)
 {
 	/*
-	 * The local apic timer can be disabled via the kernel commandline.
-	 * Register the lapic timer as a dummy clock event source on SMP
-	 * systems, so the broadcast mechanism is used. On UP systems simply
-	 * ignore it.
+	 * The local apic timer can be disabled via the kernel
+	 * commandline or from the CPU detection code. Register the lapic
+	 * timer as a dummy clock event source on SMP systems, so the
+	 * broadcast mechanism is used. On UP systems simply ignore it.
 	 */
 	if (disable_apic_timer) {
 		printk(KERN_INFO "Disabling APIC timer\n");
@@ -481,7 +506,9 @@ void __init setup_boot_APIC_clock(void)
 		return;
 	}
 
-	printk(KERN_INFO "Using local APIC timer interrupts.\n");
+	apic_printk(APIC_VERBOSE, "Using local APIC timer interrupts.\n"
+		    "calibrating APIC timer ...\n");
+
 	if (calibrate_APIC_clock()) {
 		/* No broadcast on UP ! */
 		if (num_possible_cpus() > 1)
@@ -500,6 +527,7 @@ void __init setup_boot_APIC_clock(void)
 		printk(KERN_WARNING "APIC timer registered as dummy,"
 			" due to nmi_watchdog=%d!\n", nmi_watchdog);
 
+	/* Setup the lapic or request the broadcast */
 	setup_APIC_timer();
 }
 
@@ -538,7 +566,11 @@ static void local_apic_timer_interrupt(void)
 	/*
 	 * the NMI deadlock-detector uses this.
 	 */
+#ifdef CONFIG_X86_64
 	add_pda(apic_timer_irqs, 1);
+#else
+	per_cpu(irq_stat, cpu).apic_timer_irqs++;
+#endif
 
 	evt->event_handler(evt);
 }
@@ -569,6 +601,7 @@ void smp_apic_timer_interrupt(struct pt_regs *regs)
 	irq_enter();
 	local_apic_timer_interrupt();
 	irq_exit();
+
 	set_irq_regs(old_regs);
 }
 
@@ -622,6 +655,13 @@ void clear_local_APIC(void)
 		apic_write(APIC_LVTPC, v | APIC_LVT_MASKED);
 	}
 
+	/* lets not touch this if we didn't frob it */
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL)
+	if (maxlvt >= 5) {
+		v = apic_read(APIC_LVTTHMR);
+		apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED);
+	}
+#endif
 	/*
 	 * Clean APIC state for other OSs:
 	 */
@@ -632,8 +672,14 @@ void clear_local_APIC(void)
 		apic_write(APIC_LVTERR, APIC_LVT_MASKED);
 	if (maxlvt >= 4)
 		apic_write(APIC_LVTPC, APIC_LVT_MASKED);
-	apic_write(APIC_ESR, 0);
-	apic_read(APIC_ESR);
+
+	/* Integrated APIC (!82489DX) ? */
+	if (lapic_is_integrated()) {
+		if (maxlvt > 3)
+			/* Clear ESR due to Pentium errata 3AP and 11AP */
+			apic_write(APIC_ESR, 0);
+		apic_read(APIC_ESR);
+	}
 }
 
 /**
@@ -652,8 +698,28 @@ void disable_local_APIC(void)
 	value = apic_read(APIC_SPIV);
 	value &= ~APIC_SPIV_APIC_ENABLED;
 	apic_write(APIC_SPIV, value);
+
+#ifdef CONFIG_X86_32
+	/*
+	 * When LAPIC was disabled by the BIOS and enabled by the kernel,
+	 * restore the disabled state.
+	 */
+	if (enabled_via_apicbase) {
+		unsigned int l, h;
+
+		rdmsr(MSR_IA32_APICBASE, l, h);
+		l &= ~MSR_IA32_APICBASE_ENABLE;
+		wrmsr(MSR_IA32_APICBASE, l, h);
+	}
+#endif
 }
 
+/*
+ * If Linux enabled the LAPIC against the BIOS default disable it down before
+ * re-entering the BIOS on shutdown.  Otherwise the BIOS may get confused and
+ * not power-off.  Additionally clear all LVT entries before disable_local_APIC
+ * for the case where Linux didn't enable the LAPIC.
+ */
 void lapic_shutdown(void)
 {
 	unsigned long flags;
@@ -663,7 +729,13 @@ void lapic_shutdown(void)
 
 	local_irq_save(flags);
 
-	disable_local_APIC();
+#ifdef CONFIG_X86_32
+	if (!enabled_via_apicbase)
+		clear_local_APIC();
+	else
+#endif
+		disable_local_APIC();
+
 
 	local_irq_restore(flags);
 }
@@ -734,8 +806,11 @@ int __init verify_local_APIC(void)
  */
 void __init sync_Arb_IDs(void)
 {
-	/* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 */
-	if (modern_apic())
+	/*
+	 * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not
+	 * needed on AMD.
+	 */
+	if (modern_apic() || boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
 		return;
 
 	/*
@@ -744,8 +819,8 @@ void __init sync_Arb_IDs(void)
 	apic_wait_icr_idle();
 
 	apic_printk(APIC_DEBUG, "Synchronizing Arb IDs.\n");
-	apic_write(APIC_ICR, APIC_DEST_ALLINC | APIC_INT_LEVELTRIG
-				| APIC_DM_INIT);
+	apic_write(APIC_ICR, APIC_DEST_ALLINC |
+			APIC_INT_LEVELTRIG | APIC_DM_INIT);
 }
 
 /*
@@ -762,8 +837,6 @@ void __init init_bsp_APIC(void)
 	if (smp_found_config || !cpu_has_apic)
 		return;
 
-	value = apic_read(APIC_LVR);
-
 	/*
 	 * Do not trust the local APIC being empty at bootup.
 	 */
@@ -775,7 +848,15 @@ void __init init_bsp_APIC(void)
 	value = apic_read(APIC_SPIV);
 	value &= ~APIC_VECTOR_MASK;
 	value |= APIC_SPIV_APIC_ENABLED;
-	value |= APIC_SPIV_FOCUS_DISABLED;
+
+#ifdef CONFIG_X86_32
+	/* This bit is reserved on P4/Xeon and should be cleared */
+	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
+	    (boot_cpu_data.x86 == 15))
+		value &= ~APIC_SPIV_FOCUS_DISABLED;
+	else
+#endif
+		value |= APIC_SPIV_FOCUS_DISABLED;
 	value |= SPURIOUS_APIC_VECTOR;
 	apic_write(APIC_SPIV, value);
 
@@ -784,9 +865,50 @@ void __init init_bsp_APIC(void)
 	 */
 	apic_write(APIC_LVT0, APIC_DM_EXTINT);
 	value = APIC_DM_NMI;
+	if (!lapic_is_integrated())		/* 82489DX */
+		value |= APIC_LVT_LEVEL_TRIGGER;
 	apic_write(APIC_LVT1, value);
 }
 
+static void __cpuinit lapic_setup_esr(void)
+{
+	unsigned long oldvalue, value, maxlvt;
+	if (lapic_is_integrated() && !esr_disable) {
+		if (esr_disable) {
+			/*
+			 * Something untraceable is creating bad interrupts on
+			 * secondary quads ... for the moment, just leave the
+			 * ESR disabled - we can't do anything useful with the
+			 * errors anyway - mbligh
+			 */
+			printk(KERN_INFO "Leaving ESR disabled.\n");
+			return;
+		}
+		/* !82489DX */
+		maxlvt = lapic_get_maxlvt();
+		if (maxlvt > 3)		/* Due to the Pentium erratum 3AP. */
+			apic_write(APIC_ESR, 0);
+		oldvalue = apic_read(APIC_ESR);
+
+		/* enables sending errors */
+		value = ERROR_APIC_VECTOR;
+		apic_write(APIC_LVTERR, value);
+		/*
+		 * spec says clear errors after enabling vector.
+		 */
+		if (maxlvt > 3)
+			apic_write(APIC_ESR, 0);
+		value = apic_read(APIC_ESR);
+		if (value != oldvalue)
+			apic_printk(APIC_VERBOSE, "ESR value before enabling "
+				"vector: 0x%08lx  after: 0x%08lx\n",
+				oldvalue, value);
+	} else {
+		printk(KERN_INFO "No ESR for 82489DX.\n");
+	}
+}
+
+
 /**
  * setup_local_APIC - setup the local APIC
  */
@@ -892,21 +1014,20 @@ void __cpuinit setup_local_APIC(void)
 	preempt_enable();
 }
 
-static void __cpuinit lapic_setup_esr(void)
-{
-	unsigned maxlvt = lapic_get_maxlvt();
-
-	apic_write(APIC_LVTERR, ERROR_APIC_VECTOR);
-	/*
-	 * spec says clear errors after enabling vector.
-	 */
-	if (maxlvt > 3)
-		apic_write(APIC_ESR, 0);
-}
-
 void __cpuinit end_local_APIC_setup(void)
 {
 	lapic_setup_esr();
+
+#ifdef CONFIG_X86_32
+	{
+		unsigned int value;
+		/* Disable the local apic timer */
+		value = apic_read(APIC_LVTT);
+		value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
+		apic_write(APIC_LVTT, value);
+	}
+#endif
+
 	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
 }
@@ -1108,6 +1229,8 @@ void __init init_apic_mappings(void)
  * This initializes the IO-APIC and APIC hardware if this is
  * a UP kernel.
  */
+int apic_version[MAX_APICS];
+
 int __init APIC_init_uniprocessor(void)
 {
 	if (disable_apic) {
@@ -1209,17 +1332,57 @@ asmlinkage void smp_error_interrupt(void)
 }
 
 /**
- *  * connect_bsp_APIC - attach the APIC to the interrupt system
- *   */
+ * connect_bsp_APIC - attach the APIC to the interrupt system
+ */
 void __init connect_bsp_APIC(void)
 {
+#ifdef CONFIG_X86_32
+	if (pic_mode) {
+		/*
+		 * Do not trust the local APIC being empty at bootup.
+		 */
+		clear_local_APIC();
+		/*
+		 * PIC mode, enable APIC mode in the IMCR, i.e.  connect BSP's
+		 * local APIC to INT and NMI lines.
+		 */
+		apic_printk(APIC_VERBOSE, "leaving PIC mode, "
+				"enabling APIC mode.\n");
+		outb(0x70, 0x22);
+		outb(0x01, 0x23);
+	}
+#endif
 	enable_apic_mode();
 }
 
+/**
+ * disconnect_bsp_APIC - detach the APIC from the interrupt system
+ * @virt_wire_setup:	indicates, whether virtual wire mode is selected
+ *
+ * Virtual wire mode is necessary to deliver legacy interrupts even when the
+ * APIC is disabled.
+ */
 void disconnect_bsp_APIC(int virt_wire_setup)
 {
+	unsigned int value;
+
+#ifdef CONFIG_X86_32
+	if (pic_mode) {
+		/*
+		 * Put the board back into PIC mode (has an effect only on
+		 * certain older boards).  Note that APIC interrupts, including
+		 * IPIs, won't work beyond this point!  The only exception are
+		 * INIT IPIs.
+		 */
+		apic_printk(APIC_VERBOSE, "disabling APIC mode, "
+				"entering PIC mode.\n");
+		outb(0x70, 0x22);
+		outb(0x00, 0x23);
+		return;
+	}
+#endif
+
 	/* Go back to Virtual Wire compatibility mode */
-	unsigned long value;
 
 	/* For the spurious interrupt use vector F, and enable it */
 	value = apic_read(APIC_SPIV);
@@ -1245,7 +1408,10 @@ void disconnect_bsp_APIC(int virt_wire_setup)
 		apic_write(APIC_LVT0, APIC_LVT_MASKED);
 	}
 
-	/* For LVT1 make it edge triggered, active high, nmi and enabled */
+	/*
+	 * For LVT1 make it edge triggered, active high,
+	 * nmi and enabled
+	 */
 	value = apic_read(APIC_LVT1);
 	value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
 			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
@@ -1260,9 +1426,20 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	int cpu;
 	cpumask_t tmp_map;
 
+	/*
+	 * Validate version
+	 */
+	if (version == 0x0) {
+		printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
+				"fixing up to 0x10. (tell your hw vendor)\n",
+				version);
+		version = 0x10;
+	}
+	apic_version[apicid] = version;
+
 	if (num_processors >= NR_CPUS) {
 		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
-		       " Processor ignored.\n", NR_CPUS);
+			"  Processor ignored.\n", NR_CPUS);
 		return;
 	}
 
@@ -1282,6 +1459,29 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	if (apicid > max_physical_apicid)
 		max_physical_apicid = apicid;
 
+#ifdef CONFIG_X86_32
+	/*
+	 * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
+	 * but we need to work other dependencies like SMP_SUSPEND etc
+	 * before this can be done without some confusion.
+	 * if (CPU_HOTPLUG_ENABLED || num_processors > 8)
+	 *       - Ashok Raj <ashok.raj@intel.com>
+	 */
+	if (max_physical_apicid >= 8) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_INTEL:
+			if (!APIC_XAPIC(version)) {
+				def_to_bigsmp = 0;
+				break;
+			}
+			/* If P4 and above fall through */
+		case X86_VENDOR_AMD:
+			def_to_bigsmp = 1;
+		}
+	}
+#endif
+
+#if defined(CONFIG_X86_SMP) || defined(CONFIG_X86_64)
 	/* are we being called early in kernel startup? */
 	if (early_per_cpu_ptr(x86_cpu_to_apicid)) {
 		u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);
@@ -1293,6 +1493,7 @@ void __cpuinit generic_processor_info(int apicid, int version)
 		per_cpu(x86_cpu_to_apicid, cpu) = apicid;
 		per_cpu(x86_bios_cpu_apicid, cpu) = apicid;
 	}
+#endif
 
 	cpu_set(cpu, cpu_possible_map);
 	cpu_set(cpu, cpu_present_map);
@@ -1309,9 +1510,11 @@ int hard_smp_processor_id(void)
 #ifdef CONFIG_PM
 
 static struct {
-	/* 'active' is true if the local APIC was enabled by us and
-	   not the BIOS; this signifies that we are also responsible
-	   for disabling it before entering apm/acpi suspend */
+	/*
+	 * 'active' is true if the local APIC was enabled by us and
+	 * not the BIOS; this signifies that we are also responsible
+	 * for disabling it before entering apm/acpi suspend
+	 */
 	int active;
 	/* r/w apic fields */
 	unsigned int apic_id;
@@ -1352,10 +1555,11 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 	apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR);
 	apic_pm_state.apic_tmict = apic_read(APIC_TMICT);
 	apic_pm_state.apic_tdcr = apic_read(APIC_TDCR);
-#ifdef CONFIG_X86_MCE_INTEL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR);
 #endif
+
 	local_irq_save(flags);
 	disable_local_APIC();
 	local_irq_restore(flags);
@@ -1374,13 +1578,24 @@ static int lapic_resume(struct sys_device *dev)
 	maxlvt = lapic_get_maxlvt();
 
 	local_irq_save(flags);
-	if (!x2apic) {
+
+#ifdef CONFIG_X86_64
+	if (x2apic)
+		enable_x2apic();
+	else
+#endif
+	{
+		/*
+		 * Make sure the APICBASE points to the right address
+		 *
+		 * FIXME! This will be wrong if we ever support suspend on
+		 * SMP! We'll need to do this as part of the CPU restore!
+		 */
 		rdmsr(MSR_IA32_APICBASE, l, h);
 		l &= ~MSR_IA32_APICBASE_BASE;
 		l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr;
 		wrmsr(MSR_IA32_APICBASE, l, h);
-	} else
-		enable_x2apic();
+	}
 
 	apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED);
 	apic_write(APIC_ID, apic_pm_state.apic_id);
@@ -1390,7 +1605,7 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_SPIV, apic_pm_state.apic_spiv);
 	apic_write(APIC_LVT0, apic_pm_state.apic_lvt0);
 	apic_write(APIC_LVT1, apic_pm_state.apic_lvt1);
-#ifdef CONFIG_X86_MCE_INTEL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr);
 #endif
@@ -1404,10 +1619,17 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_LVTERR, apic_pm_state.apic_lvterr);
 	apic_write(APIC_ESR, 0);
 	apic_read(APIC_ESR);
+
 	local_irq_restore(flags);
+
 	return 0;
 }
 
+/*
+ * This device has no shutdown method - fully functioning local APICs
+ * are needed on every CPU up until machine_halt/restart/poweroff.
+ */
+
 static struct sysdev_class lapic_sysclass = {
 	.name		= "lapic",
 	.resume		= lapic_resume,
@@ -1533,28 +1755,7 @@ early_param("nox2apic", setup_nox2apic);
 /*
  * APIC command line parameters
  */
-static int __init apic_set_verbosity(char *str)
-{
-	if (str == NULL)  {
-		skip_ioapic_setup = 0;
-		ioapic_force = 1;
-		return 0;
-	}
-	if (strcmp("debug", str) == 0)
-		apic_verbosity = APIC_DEBUG;
-	else if (strcmp("verbose", str) == 0)
-		apic_verbosity = APIC_VERBOSE;
-	else {
-		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
-				" use apic=verbose or apic=debug\n", str);
-		return -EINVAL;
-	}
-
-	return 0;
-}
-early_param("apic", apic_set_verbosity);
-
-static __init int setup_disableapic(char *str)
+static int __init setup_disableapic(char *arg)
 {
 	disable_apic = 1;
 	setup_clear_cpu_cap(X86_FEATURE_APIC);
@@ -1563,9 +1764,9 @@ static __init int setup_disableapic(char *str)
 early_param("disableapic", setup_disableapic);
 
 /* same as disableapic, for compatibility */
-static __init int setup_nolapic(char *str)
+static int __init setup_nolapic(char *arg)
 {
-	return setup_disableapic(str);
+	return setup_disableapic(arg);
 }
 early_param("nolapic", setup_nolapic);
 
@@ -1576,14 +1777,19 @@ static int __init parse_lapic_timer_c2_ok(char *arg)
 }
 early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok);
 
-static __init int setup_noapictimer(char *str)
+static int __init parse_disable_apic_timer(char *arg)
 {
-	if (str[0] != ' ' && str[0] != 0)
-		return 0;
 	disable_apic_timer = 1;
-	return 1;
+	return 0;
+}
+early_param("noapictimer", parse_disable_apic_timer);
+
+static int __init parse_nolapic_timer(char *arg)
+{
+	disable_apic_timer = 1;
+	return 0;
 }
-__setup("noapictimer", setup_noapictimer);
+early_param("nolapic_timer", parse_nolapic_timer);
 
 static __init int setup_apicpmtimer(char *s)
 {
@@ -1593,6 +1799,31 @@ static __init int setup_apicpmtimer(char *s)
 }
 __setup("apicpmtimer", setup_apicpmtimer);
 
+static int __init apic_set_verbosity(char *arg)
+{
+	if (!arg)  {
+#ifdef CONFIG_X86_64
+		skip_ioapic_setup = 0;
+		ioapic_force = 1;
+		return 0;
+#endif
+		return -EINVAL;
+	}
+
+	if (strcmp("debug", arg) == 0)
+		apic_verbosity = APIC_DEBUG;
+	else if (strcmp("verbose", arg) == 0)
+		apic_verbosity = APIC_VERBOSE;
+	else {
+		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
+			" use apic=verbose or apic=debug\n", arg);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+early_param("apic", apic_set_verbosity);
+
 static int __init lapic_insert_resource(void)
 {
 	if (!apic_phys)
diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index 9ee24e6bc4b..5145a6e72bb 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -228,12 +228,12 @@
 #include <linux/suspend.h>
 #include <linux/kthread.h>
 #include <linux/jiffies.h>
-#include <linux/smp_lock.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/desc.h>
 #include <asm/i8253.h>
+#include <asm/olpc.h>
 #include <asm/paravirt.h>
 #include <asm/reboot.h>
 
@@ -2217,7 +2217,7 @@ static int __init apm_init(void)
 
 	dmi_check_system(apm_dmi_table);
 
-	if (apm_info.bios.version == 0 || paravirt_enabled()) {
+	if (apm_info.bios.version == 0 || paravirt_enabled() || machine_is_olpc()) {
 		printk(KERN_INFO "apm: BIOS not found.\n");
 		return -ENODEV;
 	}
diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c
index c639bd55391..fdd585f9c53 100644
--- a/arch/x86/kernel/bios_uv.c
+++ b/arch/x86/kernel/bios_uv.c
@@ -25,11 +25,11 @@ x86_bios_strerror(long status)
 {
 	const char *str;
 	switch (status) {
-	case  0: str = "Call completed without error"; break;
-	case -1: str = "Not implemented"; break;
-	case -2: str = "Invalid argument"; break;
-	case -3: str = "Call completed with error"; break;
-	default: str = "Unknown BIOS status code"; break;
+	case  0: str = "Call completed without error";	break;
+	case -1: str = "Not implemented";		break;
+	case -2: str = "Invalid argument";		break;
+	case -3: str = "Call completed with error";	break;
+	default: str = "Unknown BIOS status code";	break;
 	}
 	return str;
 }
diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index 403e689df0b..7f0b45a5d78 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -3,15 +3,13 @@
 #
 
 obj-y			:= intel_cacheinfo.o addon_cpuid_features.o
-obj-y			+= proc.o capflags.o powerflags.o
+obj-y			+= proc.o capflags.o powerflags.o common.o
 
-obj-$(CONFIG_X86_32)	+= common.o bugs.o cmpxchg.o
-obj-$(CONFIG_X86_64)	+= common_64.o bugs_64.o
+obj-$(CONFIG_X86_32)	+= bugs.o cmpxchg.o
+obj-$(CONFIG_X86_64)	+= bugs_64.o
 
-obj-$(CONFIG_CPU_SUP_INTEL_32)		+= intel.o
-obj-$(CONFIG_CPU_SUP_INTEL_64)		+= intel_64.o
-obj-$(CONFIG_CPU_SUP_AMD_32)		+= amd.o
-obj-$(CONFIG_CPU_SUP_AMD_64)		+= amd_64.o
+obj-$(CONFIG_CPU_SUP_INTEL)		+= intel.o
+obj-$(CONFIG_CPU_SUP_AMD)		+= amd.o
 obj-$(CONFIG_CPU_SUP_CYRIX_32)		+= cyrix.o
 obj-$(CONFIG_CPU_SUP_CENTAUR_32)	+= centaur.o
 obj-$(CONFIG_CPU_SUP_CENTAUR_64)	+= centaur_64.o
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index d64ea6097ca..32e73520adf 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -1,13 +1,22 @@
 #include <linux/init.h>
 #include <linux/bitops.h>
 #include <linux/mm.h>
+
 #include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/apic.h>
 
+#ifdef CONFIG_X86_64
+# include <asm/numa_64.h>
+# include <asm/mmconfig.h>
+# include <asm/cacheflush.h>
+#endif
+
 #include <mach_apic.h>
+
 #include "cpu.h"
 
+#ifdef CONFIG_X86_32
 /*
  *	B step AMD K6 before B 9730xxxx have hardware bugs that can cause
  *	misexecution of code under Linux. Owners of such processors should
@@ -24,26 +33,273 @@
 extern void vide(void);
 __asm__(".align 4\nvide: ret");
 
-static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
+static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c)
 {
-	if (cpuid_eax(0x80000000) >= 0x80000007) {
-		c->x86_power = cpuid_edx(0x80000007);
-		if (c->x86_power & (1<<8))
-			set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+/*
+ * General Systems BIOSen alias the cpu frequency registers
+ * of the Elan at 0x000df000. Unfortuantly, one of the Linux
+ * drivers subsequently pokes it, and changes the CPU speed.
+ * Workaround : Remove the unneeded alias.
+ */
+#define CBAR		(0xfffc) /* Configuration Base Address  (32-bit) */
+#define CBAR_ENB	(0x80000000)
+#define CBAR_KEY	(0X000000CB)
+	if (c->x86_model == 9 || c->x86_model == 10) {
+		if (inl (CBAR) & CBAR_ENB)
+			outl (0 | CBAR_KEY, CBAR);
 	}
-
-	/*  Set MTRR capability flag if appropriate */
-	if (c->x86_model == 13 || c->x86_model == 9 ||
-	   (c->x86_model == 8 && c->x86_mask >= 8))
-		set_cpu_cap(c, X86_FEATURE_K6_MTRR);
 }
 
-static void __cpuinit init_amd(struct cpuinfo_x86 *c)
+
+static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	int mbytes = num_physpages >> (20-PAGE_SHIFT);
-	int r;
 
+	if (c->x86_model < 6) {
+		/* Based on AMD doc 20734R - June 2000 */
+		if (c->x86_model == 0) {
+			clear_cpu_cap(c, X86_FEATURE_APIC);
+			set_cpu_cap(c, X86_FEATURE_PGE);
+		}
+		return;
+	}
+
+	if (c->x86_model == 6 && c->x86_mask == 1) {
+		const int K6_BUG_LOOP = 1000000;
+		int n;
+		void (*f_vide)(void);
+		unsigned long d, d2;
+
+		printk(KERN_INFO "AMD K6 stepping B detected - ");
+
+		/*
+		 * It looks like AMD fixed the 2.6.2 bug and improved indirect
+		 * calls at the same time.
+		 */
+
+		n = K6_BUG_LOOP;
+		f_vide = vide;
+		rdtscl(d);
+		while (n--)
+			f_vide();
+		rdtscl(d2);
+		d = d2-d;
+
+		if (d > 20*K6_BUG_LOOP)
+			printk("system stability may be impaired when more than 32 MB are used.\n");
+		else
+			printk("probably OK (after B9730xxxx).\n");
+		printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
+	}
+
+	/* K6 with old style WHCR */
+	if (c->x86_model < 8 ||
+	   (c->x86_model == 8 && c->x86_mask < 8)) {
+		/* We can only write allocate on the low 508Mb */
+		if (mbytes > 508)
+			mbytes = 508;
+
+		rdmsr(MSR_K6_WHCR, l, h);
+		if ((l&0x0000FFFF) == 0) {
+			unsigned long flags;
+			l = (1<<0)|((mbytes/4)<<1);
+			local_irq_save(flags);
+			wbinvd();
+			wrmsr(MSR_K6_WHCR, l, h);
+			local_irq_restore(flags);
+			printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n",
+				mbytes);
+		}
+		return;
+	}
+
+	if ((c->x86_model == 8 && c->x86_mask > 7) ||
+	     c->x86_model == 9 || c->x86_model == 13) {
+		/* The more serious chips .. */
+
+		if (mbytes > 4092)
+			mbytes = 4092;
+
+		rdmsr(MSR_K6_WHCR, l, h);
+		if ((l&0xFFFF0000) == 0) {
+			unsigned long flags;
+			l = ((mbytes>>2)<<22)|(1<<16);
+			local_irq_save(flags);
+			wbinvd();
+			wrmsr(MSR_K6_WHCR, l, h);
+			local_irq_restore(flags);
+			printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n",
+				mbytes);
+		}
+
+		return;
+	}
+
+	if (c->x86_model == 10) {
+		/* AMD Geode LX is model 10 */
+		/* placeholder for any needed mods */
+		return;
+	}
+}
+
+static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c)
+{
+	u32 l, h;
+
+	/*
+	 * Bit 15 of Athlon specific MSR 15, needs to be 0
+	 * to enable SSE on Palomino/Morgan/Barton CPU's.
+	 * If the BIOS didn't enable it already, enable it here.
+	 */
+	if (c->x86_model >= 6 && c->x86_model <= 10) {
+		if (!cpu_has(c, X86_FEATURE_XMM)) {
+			printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
+			rdmsr(MSR_K7_HWCR, l, h);
+			l &= ~0x00008000;
+			wrmsr(MSR_K7_HWCR, l, h);
+			set_cpu_cap(c, X86_FEATURE_XMM);
+		}
+	}
+
+	/*
+	 * It's been determined by AMD that Athlons since model 8 stepping 1
+	 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
+	 * As per AMD technical note 27212 0.2
+	 */
+	if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
+		rdmsr(MSR_K7_CLK_CTL, l, h);
+		if ((l & 0xfff00000) != 0x20000000) {
+			printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
+				((l & 0x000fffff)|0x20000000));
+			wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
+		}
+	}
+
+	set_cpu_cap(c, X86_FEATURE_K7);
+}
+#endif
+
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+static int __cpuinit nearby_node(int apicid)
+{
+	int i, node;
+
+	for (i = apicid - 1; i >= 0; i--) {
+		node = apicid_to_node[i];
+		if (node != NUMA_NO_NODE && node_online(node))
+			return node;
+	}
+	for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
+		node = apicid_to_node[i];
+		if (node != NUMA_NO_NODE && node_online(node))
+			return node;
+	}
+	return first_node(node_online_map); /* Shouldn't happen */
+}
+#endif
+
+/*
+ * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
+ * Assumes number of cores is a power of two.
+ */
+static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_HT
+	unsigned bits;
+
+	bits = c->x86_coreid_bits;
+
+	/* Low order bits define the core id (index of core in socket) */
+	c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
+	/* Convert the initial APIC ID into the socket ID */
+	c->phys_proc_id = c->initial_apicid >> bits;
+#endif
+}
+
+static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
+{
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+	int cpu = smp_processor_id();
+	int node;
+	unsigned apicid = hard_smp_processor_id();
+
+	node = c->phys_proc_id;
+	if (apicid_to_node[apicid] != NUMA_NO_NODE)
+		node = apicid_to_node[apicid];
+	if (!node_online(node)) {
+		/* Two possibilities here:
+		   - The CPU is missing memory and no node was created.
+		   In that case try picking one from a nearby CPU
+		   - The APIC IDs differ from the HyperTransport node IDs
+		   which the K8 northbridge parsing fills in.
+		   Assume they are all increased by a constant offset,
+		   but in the same order as the HT nodeids.
+		   If that doesn't result in a usable node fall back to the
+		   path for the previous case.  */
+
+		int ht_nodeid = c->initial_apicid;
+
+		if (ht_nodeid >= 0 &&
+		    apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
+			node = apicid_to_node[ht_nodeid];
+		/* Pick a nearby node */
+		if (!node_online(node))
+			node = nearby_node(apicid);
+	}
+	numa_set_node(cpu, node);
+
+	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
+#endif
+}
+
+static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_HT
+	unsigned bits, ecx;
+
+	/* Multi core CPU? */
+	if (c->extended_cpuid_level < 0x80000008)
+		return;
+
+	ecx = cpuid_ecx(0x80000008);
+
+	c->x86_max_cores = (ecx & 0xff) + 1;
+
+	/* CPU telling us the core id bits shift? */
+	bits = (ecx >> 12) & 0xF;
+
+	/* Otherwise recompute */
+	if (bits == 0) {
+		while ((1 << bits) < c->x86_max_cores)
+			bits++;
+	}
+
+	c->x86_coreid_bits = bits;
+#endif
+}
+
+static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
+{
+	early_init_amd_mc(c);
+
+	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
+	if (c->x86_power & (1<<8))
+		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+
+#ifdef CONFIG_X86_64
+	set_cpu_cap(c, X86_FEATURE_SYSCALL32);
+#else
+	/*  Set MTRR capability flag if appropriate */
+	if (c->x86 == 5)
+		if (c->x86_model == 13 || c->x86_model == 9 ||
+		    (c->x86_model == 8 && c->x86_mask >= 8))
+			set_cpu_cap(c, X86_FEATURE_K6_MTRR);
+#endif
+}
+
+static void __cpuinit init_amd(struct cpuinfo_x86 *c)
+{
 #ifdef CONFIG_SMP
 	unsigned long long value;
 
@@ -54,7 +310,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	 * Errata 63 for SH-B3 steppings
 	 * Errata 122 for all steppings (F+ have it disabled by default)
 	 */
-	if (c->x86 == 15) {
+	if (c->x86 == 0xf) {
 		rdmsrl(MSR_K7_HWCR, value);
 		value |= 1 << 6;
 		wrmsrl(MSR_K7_HWCR, value);
@@ -64,209 +320,119 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	early_init_amd(c);
 
 	/*
-	 *	FIXME: We should handle the K5 here. Set up the write
-	 *	range and also turn on MSR 83 bits 4 and 31 (write alloc,
-	 *	no bus pipeline)
-	 */
-
-	/*
 	 * Bit 31 in normal CPUID used for nonstandard 3DNow ID;
 	 * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway
 	 */
 	clear_cpu_cap(c, 0*32+31);
 
-	r = get_model_name(c);
+#ifdef CONFIG_X86_64
+	/* On C+ stepping K8 rep microcode works well for copy/memset */
+	if (c->x86 == 0xf) {
+		u32 level;
 
-	switch (c->x86) {
-	case 4:
-		/*
-		 * General Systems BIOSen alias the cpu frequency registers
-		 * of the Elan at 0x000df000. Unfortuantly, one of the Linux
-		 * drivers subsequently pokes it, and changes the CPU speed.
-		 * Workaround : Remove the unneeded alias.
-		 */
-#define CBAR		(0xfffc) /* Configuration Base Address  (32-bit) */
-#define CBAR_ENB	(0x80000000)
-#define CBAR_KEY	(0X000000CB)
-			if (c->x86_model == 9 || c->x86_model == 10) {
-				if (inl (CBAR) & CBAR_ENB)
-					outl (0 | CBAR_KEY, CBAR);
-			}
-			break;
-	case 5:
-			if (c->x86_model < 6) {
-				/* Based on AMD doc 20734R - June 2000 */
-				if (c->x86_model == 0) {
-					clear_cpu_cap(c, X86_FEATURE_APIC);
-					set_cpu_cap(c, X86_FEATURE_PGE);
-				}
-				break;
-			}
-
-			if (c->x86_model == 6 && c->x86_mask == 1) {
-				const int K6_BUG_LOOP = 1000000;
-				int n;
-				void (*f_vide)(void);
-				unsigned long d, d2;
-
-				printk(KERN_INFO "AMD K6 stepping B detected - ");
-
-				/*
-				 * It looks like AMD fixed the 2.6.2 bug and improved indirect
-				 * calls at the same time.
-				 */
-
-				n = K6_BUG_LOOP;
-				f_vide = vide;
-				rdtscl(d);
-				while (n--)
-					f_vide();
-				rdtscl(d2);
-				d = d2-d;
-
-				if (d > 20*K6_BUG_LOOP)
-					printk("system stability may be impaired when more than 32 MB are used.\n");
-				else
-					printk("probably OK (after B9730xxxx).\n");
-				printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
-			}
-
-			/* K6 with old style WHCR */
-			if (c->x86_model < 8 ||
-			   (c->x86_model == 8 && c->x86_mask < 8)) {
-				/* We can only write allocate on the low 508Mb */
-				if (mbytes > 508)
-					mbytes = 508;
-
-				rdmsr(MSR_K6_WHCR, l, h);
-				if ((l&0x0000FFFF) == 0) {
-					unsigned long flags;
-					l = (1<<0)|((mbytes/4)<<1);
-					local_irq_save(flags);
-					wbinvd();
-					wrmsr(MSR_K6_WHCR, l, h);
-					local_irq_restore(flags);
-					printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n",
-						mbytes);
-				}
-				break;
-			}
-
-			if ((c->x86_model == 8 && c->x86_mask > 7) ||
-			     c->x86_model == 9 || c->x86_model == 13) {
-				/* The more serious chips .. */
-
-				if (mbytes > 4092)
-					mbytes = 4092;
-
-				rdmsr(MSR_K6_WHCR, l, h);
-				if ((l&0xFFFF0000) == 0) {
-					unsigned long flags;
-					l = ((mbytes>>2)<<22)|(1<<16);
-					local_irq_save(flags);
-					wbinvd();
-					wrmsr(MSR_K6_WHCR, l, h);
-					local_irq_restore(flags);
-					printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n",
-						mbytes);
-				}
-
-				break;
-			}
-
-			if (c->x86_model == 10) {
-				/* AMD Geode LX is model 10 */
-				/* placeholder for any needed mods */
-				break;
-			}
-			break;
-	case 6: /* An Athlon/Duron */
-
-			/*
-			 * Bit 15 of Athlon specific MSR 15, needs to be 0
-			 * to enable SSE on Palomino/Morgan/Barton CPU's.
-			 * If the BIOS didn't enable it already, enable it here.
-			 */
-			if (c->x86_model >= 6 && c->x86_model <= 10) {
-				if (!cpu_has(c, X86_FEATURE_XMM)) {
-					printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
-					rdmsr(MSR_K7_HWCR, l, h);
-					l &= ~0x00008000;
-					wrmsr(MSR_K7_HWCR, l, h);
-					set_cpu_cap(c, X86_FEATURE_XMM);
-				}
-			}
-
-			/*
-			 * It's been determined by AMD that Athlons since model 8 stepping 1
-			 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
-			 * As per AMD technical note 27212 0.2
-			 */
-			if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
-				rdmsr(MSR_K7_CLK_CTL, l, h);
-				if ((l & 0xfff00000) != 0x20000000) {
-					printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
-						((l & 0x000fffff)|0x20000000));
-					wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
-				}
-			}
-			break;
+		level = cpuid_eax(1);
+		if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
+			set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	}
+	if (c->x86 == 0x10 || c->x86 == 0x11)
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+#else
+
+	/*
+	 *	FIXME: We should handle the K5 here. Set up the write
+	 *	range and also turn on MSR 83 bits 4 and 31 (write alloc,
+	 *	no bus pipeline)
+	 */
 
 	switch (c->x86) {
-	case 15:
-	/* Use K8 tuning for Fam10h and Fam11h */
-	case 0x10:
-	case 0x11:
-		set_cpu_cap(c, X86_FEATURE_K8);
+	case 4:
+		init_amd_k5(c);
 		break;
-	case 6:
-		set_cpu_cap(c, X86_FEATURE_K7);
+	case 5:
+		init_amd_k6(c);
+		break;
+	case 6: /* An Athlon/Duron */
+		init_amd_k7(c);
 		break;
 	}
+
+	/* K6s reports MCEs but don't actually have all the MSRs */
+	if (c->x86 < 6)
+		clear_cpu_cap(c, X86_FEATURE_MCE);
+#endif
+
+	/* Enable workaround for FXSAVE leak */
 	if (c->x86 >= 6)
 		set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK);
 
-	display_cacheinfo(c);
-
-	if (cpuid_eax(0x80000000) >= 0x80000008)
-		c->x86_max_cores = (cpuid_ecx(0x80000008) & 0xff) + 1;
+	if (!c->x86_model_id[0]) {
+		switch (c->x86) {
+		case 0xf:
+			/* Should distinguish Models here, but this is only
+			   a fallback anyways. */
+			strcpy(c->x86_model_id, "Hammer");
+			break;
+		}
+	}
 
-#ifdef CONFIG_X86_HT
-	/*
-	 * On a AMD multi core setup the lower bits of the APIC id
-	 * distinguish the cores.
-	 */
-	if (c->x86_max_cores > 1) {
-		int cpu = smp_processor_id();
-		unsigned bits = (cpuid_ecx(0x80000008) >> 12) & 0xf;
+	display_cacheinfo(c);
 
-		if (bits == 0) {
-			while ((1 << bits) < c->x86_max_cores)
-				bits++;
-		}
-		c->cpu_core_id = c->phys_proc_id & ((1<<bits)-1);
-		c->phys_proc_id >>= bits;
-		printk(KERN_INFO "CPU %d(%d) -> Core %d\n",
-		       cpu, c->x86_max_cores, c->cpu_core_id);
+	/* Multi core CPU? */
+	if (c->extended_cpuid_level >= 0x80000008) {
+		amd_detect_cmp(c);
+		srat_detect_node(c);
 	}
+
+#ifdef CONFIG_X86_32
+	detect_ht(c);
 #endif
 
-	if (cpuid_eax(0x80000000) >= 0x80000006) {
-		if ((c->x86 == 0x10) && (cpuid_edx(0x80000006) & 0xf000))
+	if (c->extended_cpuid_level >= 0x80000006) {
+		if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000))
 			num_cache_leaves = 4;
 		else
 			num_cache_leaves = 3;
 	}
 
-	/* K6s reports MCEs but don't actually have all the MSRs */
-	if (c->x86 < 6)
-		clear_cpu_cap(c, X86_FEATURE_MCE);
+	if (c->x86 >= 0xf && c->x86 <= 0x11)
+		set_cpu_cap(c, X86_FEATURE_K8);
 
-	if (cpu_has_xmm2)
+	if (cpu_has_xmm2) {
+		/* MFENCE stops RDTSC speculation */
 		set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
+	}
+
+#ifdef CONFIG_X86_64
+	if (c->x86 == 0x10) {
+		/* do this for boot cpu */
+		if (c == &boot_cpu_data)
+			check_enable_amd_mmconf_dmi();
+
+		fam10h_check_enable_mmcfg();
+	}
+
+	if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) {
+		unsigned long long tseg;
+
+		/*
+		 * Split up direct mapping around the TSEG SMM area.
+		 * Don't do it for gbpages because there seems very little
+		 * benefit in doing so.
+		 */
+		if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
+		    printk(KERN_DEBUG "tseg: %010llx\n", tseg);
+		    if ((tseg>>PMD_SHIFT) <
+				(max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
+			((tseg>>PMD_SHIFT) <
+				(max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
+			 (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
+			set_memory_4k((unsigned long)__va(tseg), 1);
+		}
+	}
+#endif
 }
 
+#ifdef CONFIG_X86_32
 static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size)
 {
 	/* AMD errata T13 (order #21922) */
@@ -279,10 +445,12 @@ static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int
 	}
 	return size;
 }
+#endif
 
 static struct cpu_dev amd_cpu_dev __cpuinitdata = {
 	.c_vendor	= "AMD",
 	.c_ident	= { "AuthenticAMD" },
+#ifdef CONFIG_X86_32
 	.c_models = {
 		{ .vendor = X86_VENDOR_AMD, .family = 4, .model_names =
 		  {
@@ -295,9 +463,10 @@ static struct cpu_dev amd_cpu_dev __cpuinitdata = {
 		  }
 		},
 	},
+	.c_size_cache	= amd_size_cache,
+#endif
 	.c_early_init   = early_init_amd,
 	.c_init		= init_amd,
-	.c_size_cache	= amd_size_cache,
 	.c_x86_vendor	= X86_VENDOR_AMD,
 };
 
diff --git a/arch/x86/kernel/cpu/amd_64.c b/arch/x86/kernel/cpu/amd_64.c
deleted file mode 100644
index d1c721c0c49..00000000000
--- a/arch/x86/kernel/cpu/amd_64.c
+++ /dev/null
@@ -1,224 +0,0 @@
-#include <linux/init.h>
-#include <linux/mm.h>
-
-#include <asm/numa_64.h>
-#include <asm/mmconfig.h>
-#include <asm/cacheflush.h>
-
-#include <mach_apic.h>
-
-#include "cpu.h"
-
-int force_mwait __cpuinitdata;
-
-#ifdef CONFIG_NUMA
-static int __cpuinit nearby_node(int apicid)
-{
-	int i, node;
-
-	for (i = apicid - 1; i >= 0; i--) {
-		node = apicid_to_node[i];
-		if (node != NUMA_NO_NODE && node_online(node))
-			return node;
-	}
-	for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
-		node = apicid_to_node[i];
-		if (node != NUMA_NO_NODE && node_online(node))
-			return node;
-	}
-	return first_node(node_online_map); /* Shouldn't happen */
-}
-#endif
-
-/*
- * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
- * Assumes number of cores is a power of two.
- */
-static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-	unsigned bits;
-#ifdef CONFIG_NUMA
-	int cpu = smp_processor_id();
-	int node = 0;
-	unsigned apicid = hard_smp_processor_id();
-#endif
-	bits = c->x86_coreid_bits;
-
-	/* Low order bits define the core id (index of core in socket) */
-	c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
-	/* Convert the initial APIC ID into the socket ID */
-	c->phys_proc_id = c->initial_apicid >> bits;
-
-#ifdef CONFIG_NUMA
-	node = c->phys_proc_id;
-	if (apicid_to_node[apicid] != NUMA_NO_NODE)
-		node = apicid_to_node[apicid];
-	if (!node_online(node)) {
-		/* Two possibilities here:
-		   - The CPU is missing memory and no node was created.
-		   In that case try picking one from a nearby CPU
-		   - The APIC IDs differ from the HyperTransport node IDs
-		   which the K8 northbridge parsing fills in.
-		   Assume they are all increased by a constant offset,
-		   but in the same order as the HT nodeids.
-		   If that doesn't result in a usable node fall back to the
-		   path for the previous case.  */
-
-		int ht_nodeid = c->initial_apicid;
-
-		if (ht_nodeid >= 0 &&
-		    apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
-			node = apicid_to_node[ht_nodeid];
-		/* Pick a nearby node */
-		if (!node_online(node))
-			node = nearby_node(apicid);
-	}
-	numa_set_node(cpu, node);
-
-	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
-#endif
-#endif
-}
-
-static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-	unsigned bits, ecx;
-
-	/* Multi core CPU? */
-	if (c->extended_cpuid_level < 0x80000008)
-		return;
-
-	ecx = cpuid_ecx(0x80000008);
-
-	c->x86_max_cores = (ecx & 0xff) + 1;
-
-	/* CPU telling us the core id bits shift? */
-	bits = (ecx >> 12) & 0xF;
-
-	/* Otherwise recompute */
-	if (bits == 0) {
-		while ((1 << bits) < c->x86_max_cores)
-			bits++;
-	}
-
-	c->x86_coreid_bits = bits;
-
-#endif
-}
-
-static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
-{
-	early_init_amd_mc(c);
-
-	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
-	if (c->x86_power & (1<<8))
-		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-
-	set_cpu_cap(c, X86_FEATURE_SYSCALL32);
-}
-
-static void __cpuinit init_amd(struct cpuinfo_x86 *c)
-{
-	unsigned level;
-
-#ifdef CONFIG_SMP
-	unsigned long value;
-
-	/*
-	 * Disable TLB flush filter by setting HWCR.FFDIS on K8
-	 * bit 6 of msr C001_0015
-	 *
-	 * Errata 63 for SH-B3 steppings
-	 * Errata 122 for all steppings (F+ have it disabled by default)
-	 */
-	if (c->x86 == 0xf) {
-		rdmsrl(MSR_K8_HWCR, value);
-		value |= 1 << 6;
-		wrmsrl(MSR_K8_HWCR, value);
-	}
-#endif
-
-	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
-	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
-	clear_cpu_cap(c, 0*32+31);
-
-	/* On C+ stepping K8 rep microcode works well for copy/memset */
-	if (c->x86 == 0xf) {
-		level = cpuid_eax(1);
-		if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
-			set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-	}
-	if (c->x86 == 0x10 || c->x86 == 0x11)
-		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-
-	/* Enable workaround for FXSAVE leak */
-	if (c->x86 >= 6)
-		set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK);
-
-	level = get_model_name(c);
-	if (!level) {
-		switch (c->x86) {
-		case 0xf:
-			/* Should distinguish Models here, but this is only
-			   a fallback anyways. */
-			strcpy(c->x86_model_id, "Hammer");
-			break;
-		}
-	}
-	display_cacheinfo(c);
-
-	/* Multi core CPU? */
-	if (c->extended_cpuid_level >= 0x80000008)
-		amd_detect_cmp(c);
-
-	if (c->extended_cpuid_level >= 0x80000006 &&
-		(cpuid_edx(0x80000006) & 0xf000))
-		num_cache_leaves = 4;
-	else
-		num_cache_leaves = 3;
-
-	if (c->x86 >= 0xf && c->x86 <= 0x11)
-		set_cpu_cap(c, X86_FEATURE_K8);
-
-	/* MFENCE stops RDTSC speculation */
-	set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
-
-	if (c->x86 == 0x10) {
-		/* do this for boot cpu */
-		if (c == &boot_cpu_data)
-			check_enable_amd_mmconf_dmi();
-
-		fam10h_check_enable_mmcfg();
-	}
-
-	if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) {
-		unsigned long long tseg;
-
-		/*
-		 * Split up direct mapping around the TSEG SMM area.
-		 * Don't do it for gbpages because there seems very little
-		 * benefit in doing so.
-		 */
-		if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
-		    printk(KERN_DEBUG "tseg: %010llx\n", tseg);
-		    if ((tseg>>PMD_SHIFT) <
-				(max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
-			((tseg>>PMD_SHIFT) <
-				(max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
-			 (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
-			set_memory_4k((unsigned long)__va(tseg), 1);
-		}
-	}
-}
-
-static struct cpu_dev amd_cpu_dev __cpuinitdata = {
-	.c_vendor	= "AMD",
-	.c_ident	= { "AuthenticAMD" },
-	.c_early_init   = early_init_amd,
-	.c_init		= init_amd,
-	.c_x86_vendor	= X86_VENDOR_AMD,
-};
-
-cpu_dev_register(amd_cpu_dev);
diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c
index e5f6d89521b..89bfdd9cacc 100644
--- a/arch/x86/kernel/cpu/centaur.c
+++ b/arch/x86/kernel/cpu/centaur.c
@@ -289,7 +289,6 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c)
 	if (c->x86_model >= 6 && c->x86_model < 9)
 		set_cpu_cap(c, X86_FEATURE_3DNOW);
 
-	get_model_name(c);
 	display_cacheinfo(c);
 }
 
diff --git a/arch/x86/kernel/cpu/centaur_64.c b/arch/x86/kernel/cpu/centaur_64.c
index 49cfc6d2f2f..a1625f5a1e7 100644
--- a/arch/x86/kernel/cpu/centaur_64.c
+++ b/arch/x86/kernel/cpu/centaur_64.c
@@ -16,9 +16,10 @@ static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c)
 
 static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 {
+	early_init_centaur(c);
+
 	if (c->x86 == 0x6 && c->x86_model >= 0xf) {
 		c->x86_cache_alignment = c->x86_clflush_size * 2;
-		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
 		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	}
 	set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 7d5a07f0fd2..7581b62df18 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1,29 +1,61 @@
 #include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
 #include <linux/string.h>
+#include <linux/bootmem.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/kgdb.h>
+#include <linux/topology.h>
 #include <linux/delay.h>
 #include <linux/smp.h>
-#include <linux/module.h>
 #include <linux/percpu.h>
-#include <linux/bootmem.h>
-#include <asm/processor.h>
 #include <asm/i387.h>
 #include <asm/msr.h>
 #include <asm/io.h>
+#include <asm/linkage.h>
 #include <asm/mmu_context.h>
 #include <asm/mtrr.h>
 #include <asm/mce.h>
 #include <asm/pat.h>
 #include <asm/asm.h>
+#include <asm/numa.h>
 #ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/mpspec.h>
 #include <asm/apic.h>
 #include <mach_apic.h>
+#include <asm/genapic.h>
 #endif
 
+#include <asm/pda.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/desc.h>
+#include <asm/atomic.h>
+#include <asm/proto.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+
 #include "cpu.h"
 
 static struct cpu_dev *this_cpu __cpuinitdata;
 
+#ifdef CONFIG_X86_64
+/* We need valid kernel segments for data and code in long mode too
+ * IRET will check the segment types  kkeil 2000/10/28
+ * Also sysret mandates a special GDT layout
+ */
+/* The TLS descriptors are currently at a different place compared to i386.
+   Hopefully nobody expects them at a fixed place (Wine?) */
+DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
+	[GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } },
+	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } },
+	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } },
+	[GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } },
+	[GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } },
+	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } },
+} };
+#else
 DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } },
 	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } },
@@ -58,8 +90,10 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 	[GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } },
 	[GDT_ENTRY_PERCPU] = { { { 0x00000000, 0x00000000 } } },
 } };
+#endif
 EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
+#ifdef CONFIG_X86_32
 static int cachesize_override __cpuinitdata = -1;
 static int disable_x86_serial_nr __cpuinitdata = 1;
 
@@ -70,34 +104,6 @@ static int __init cachesize_setup(char *str)
 }
 __setup("cachesize=", cachesize_setup);
 
-/*
- * Naming convention should be: <Name> [(<Codename>)]
- * This table only is used unless init_<vendor>() below doesn't set it;
- * in particular, if CPUID levels 0x80000002..4 are supported, this isn't used
- *
- */
-
-/* Look up CPU names by table lookup. */
-static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
-{
-	struct cpu_model_info *info;
-
-	if (c->x86_model >= 16)
-		return NULL;	/* Range check */
-
-	if (!this_cpu)
-		return NULL;
-
-	info = this_cpu->c_models;
-
-	while (info && info->family) {
-		if (info->family == c->x86)
-			return info->model_names[c->x86_model];
-		info++;
-	}
-	return NULL;		/* Not found */
-}
-
 static int __init x86_fxsr_setup(char *s)
 {
 	setup_clear_cpu_cap(X86_FEATURE_FXSR);
@@ -162,6 +168,48 @@ static int __init x86_serial_nr_setup(char *s)
 	return 1;
 }
 __setup("serialnumber", x86_serial_nr_setup);
+#else
+static inline int flag_is_changeable_p(u32 flag)
+{
+	return 1;
+}
+/* Probe for the CPUID instruction */
+static inline int have_cpuid_p(void)
+{
+	return 1;
+}
+static inline void squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+{
+}
+#endif
+
+/*
+ * Naming convention should be: <Name> [(<Codename>)]
+ * This table only is used unless init_<vendor>() below doesn't set it;
+ * in particular, if CPUID levels 0x80000002..4 are supported, this isn't used
+ *
+ */
+
+/* Look up CPU names by table lookup. */
+static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
+{
+	struct cpu_model_info *info;
+
+	if (c->x86_model >= 16)
+		return NULL;	/* Range check */
+
+	if (!this_cpu)
+		return NULL;
+
+	info = this_cpu->c_models;
+
+	while (info && info->family) {
+		if (info->family == c->x86)
+			return info->model_names[c->x86_model];
+		info++;
+	}
+	return NULL;		/* Not found */
+}
 
 __u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
 
@@ -174,13 +222,18 @@ void switch_to_new_gdt(void)
 	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
 	gdt_descr.size = GDT_SIZE - 1;
 	load_gdt(&gdt_descr);
+#ifdef CONFIG_X86_32
 	asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory");
+#endif
 }
 
 static struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
 
 static void __cpuinit default_init(struct cpuinfo_x86 *c)
 {
+#ifdef CONFIG_X86_64
+	display_cacheinfo(c);
+#else
 	/* Not much we can do here... */
 	/* Check if at least it has cpuid */
 	if (c->cpuid_level == -1) {
@@ -190,6 +243,7 @@ static void __cpuinit default_init(struct cpuinfo_x86 *c)
 		else if (c->x86 == 3)
 			strcpy(c->x86_model_id, "386");
 	}
+#endif
 }
 
 static struct cpu_dev __cpuinitdata default_cpu = {
@@ -198,13 +252,13 @@ static struct cpu_dev __cpuinitdata default_cpu = {
 	.c_x86_vendor = X86_VENDOR_UNKNOWN,
 };
 
-int __cpuinit get_model_name(struct cpuinfo_x86 *c)
+static void __cpuinit get_model_name(struct cpuinfo_x86 *c)
 {
 	unsigned int *v;
 	char *p, *q;
 
 	if (c->extended_cpuid_level < 0x80000004)
-		return 0;
+		return;
 
 	v = (unsigned int *) c->x86_model_id;
 	cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
@@ -223,8 +277,6 @@ int __cpuinit get_model_name(struct cpuinfo_x86 *c)
 	     while (q <= &c->x86_model_id[48])
 		  *q++ = '\0';	/* Zero-pad the rest */
 	}
-
-	return 1;
 }
 
 void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
@@ -238,6 +290,10 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 		printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
 				edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
 		c->x86_cache_size = (ecx>>24) + (edx>>24);
+#ifdef CONFIG_X86_64
+		/* On K8 L1 TLB is inclusive, so don't count it */
+		c->x86_tlbsize = 0;
+#endif
 	}
 
 	if (n < 0x80000006)	/* Some chips just has a large L1. */
@@ -246,6 +302,9 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 	cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
 	l2size = ecx >> 16;
 
+#ifdef CONFIG_X86_64
+	c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
+#else
 	/* do processor-specific cache resizing */
 	if (this_cpu->c_size_cache)
 		l2size = this_cpu->c_size_cache(c, l2size);
@@ -256,6 +315,7 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 
 	if (l2size == 0)
 		return;		/* Again, no L2 cache is possible */
+#endif
 
 	c->x86_cache_size = l2size;
 
@@ -263,9 +323,9 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 			l2size, ecx & 0xFF);
 }
 
-#ifdef CONFIG_X86_HT
 void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 {
+#ifdef CONFIG_X86_HT
 	u32 eax, ebx, ecx, edx;
 	int index_msb, core_bits;
 
@@ -275,6 +335,9 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 	if (cpu_has(c, X86_FEATURE_CMP_LEGACY))
 		goto out;
 
+	if (cpu_has(c, X86_FEATURE_XTOPOLOGY))
+		return;
+
 	cpuid(1, &eax, &ebx, &ecx, &edx);
 
 	smp_num_siblings = (ebx & 0xff0000) >> 16;
@@ -291,8 +354,11 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 		}
 
 		index_msb = get_count_order(smp_num_siblings);
+#ifdef CONFIG_X86_64
+		c->phys_proc_id = phys_pkg_id(index_msb);
+#else
 		c->phys_proc_id = phys_pkg_id(c->initial_apicid, index_msb);
-
+#endif
 
 		smp_num_siblings = smp_num_siblings / c->x86_max_cores;
 
@@ -300,8 +366,13 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 
 		core_bits = get_count_order(c->x86_max_cores);
 
+#ifdef CONFIG_X86_64
+		c->cpu_core_id = phys_pkg_id(index_msb) &
+					       ((1 << core_bits) - 1);
+#else
 		c->cpu_core_id = phys_pkg_id(c->initial_apicid, index_msb) &
 					       ((1 << core_bits) - 1);
+#endif
 	}
 
 out:
@@ -311,8 +382,8 @@ out:
 		printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
 		       c->cpu_core_id);
 	}
-}
 #endif
+}
 
 static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
 {
@@ -335,7 +406,7 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
 
 	if (!printed) {
 		printed++;
-		printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
+		printk(KERN_ERR "CPU: vendor_id '%s' unknown, using generic init.\n", v);
 		printk(KERN_ERR "CPU: Your system may be unstable.\n");
 	}
 
@@ -392,7 +463,47 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c)
 			c->x86_capability[6] = cpuid_ecx(0x80000001);
 		}
 	}
+
+#ifdef CONFIG_X86_64
+	if (c->extended_cpuid_level >= 0x80000008) {
+		u32 eax = cpuid_eax(0x80000008);
+
+		c->x86_virt_bits = (eax >> 8) & 0xff;
+		c->x86_phys_bits = eax & 0xff;
+	}
+#endif
+
+	if (c->extended_cpuid_level >= 0x80000007)
+		c->x86_power = cpuid_edx(0x80000007);
+
 }
+
+static void __cpuinit identify_cpu_without_cpuid(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_32
+	int i;
+
+	/*
+	 * First of all, decide if this is a 486 or higher
+	 * It's a 486 if we can modify the AC flag
+	 */
+	if (flag_is_changeable_p(X86_EFLAGS_AC))
+		c->x86 = 4;
+	else
+		c->x86 = 3;
+
+	for (i = 0; i < X86_VENDOR_NUM; i++)
+		if (cpu_devs[i] && cpu_devs[i]->c_identify) {
+			c->x86_vendor_id[0] = 0;
+			cpu_devs[i]->c_identify(c);
+			if (c->x86_vendor_id[0]) {
+				get_cpu_vendor(c);
+				break;
+			}
+		}
+#endif
+}
+
 /*
  * Do minimum CPU detection early.
  * Fields really needed: vendor, cpuid_level, family, model, mask,
@@ -404,16 +515,23 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c)
  */
 static void __init early_identify_cpu(struct cpuinfo_x86 *c)
 {
+#ifdef CONFIG_X86_64
+	c->x86_clflush_size = 64;
+#else
 	c->x86_clflush_size = 32;
+#endif
 	c->x86_cache_alignment = c->x86_clflush_size;
 
-	if (!have_cpuid_p())
-		return;
-
 	memset(&c->x86_capability, 0, sizeof c->x86_capability);
-
 	c->extended_cpuid_level = 0;
 
+	if (!have_cpuid_p())
+		identify_cpu_without_cpuid(c);
+
+	/* cyrix could have cpuid enabled via c_identify()*/
+	if (!have_cpuid_p())
+		return;
+
 	cpu_detect(c);
 
 	get_cpu_vendor(c);
@@ -454,39 +572,27 @@ void __init early_cpu_init(void)
 
 /*
  * The NOPL instruction is supposed to exist on all CPUs with
- * family >= 6, unfortunately, that's not true in practice because
+ * family >= 6; unfortunately, that's not true in practice because
  * of early VIA chips and (more importantly) broken virtualizers that
- * are not easy to detect.  Hence, probe for it based on first
- * principles.
+ * are not easy to detect.  In the latter case it doesn't even *fail*
+ * reliably, so probing for it doesn't even work.  Disable it completely
+ * unless we can find a reliable way to detect all the broken cases.
  */
 static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
 {
-	const u32 nopl_signature = 0x888c53b1; /* Random number */
-	u32 has_nopl = nopl_signature;
-
 	clear_cpu_cap(c, X86_FEATURE_NOPL);
-	if (c->x86 >= 6) {
-		asm volatile("\n"
-			     "1:      .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */
-			     "2:\n"
-			     "        .section .fixup,\"ax\"\n"
-			     "3:      xor %0,%0\n"
-			     "        jmp 2b\n"
-			     "        .previous\n"
-			     _ASM_EXTABLE(1b,3b)
-			     : "+a" (has_nopl));
-
-		if (has_nopl == nopl_signature)
-			set_cpu_cap(c, X86_FEATURE_NOPL);
-	}
 }
 
 static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
 {
+	c->extended_cpuid_level = 0;
+
 	if (!have_cpuid_p())
-		return;
+		identify_cpu_without_cpuid(c);
 
-	c->extended_cpuid_level = 0;
+	/* cyrix could have cpuid enabled via c_identify()*/
+	if (!have_cpuid_p())
+		return;
 
 	cpu_detect(c);
 
@@ -496,16 +602,20 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
 
 	if (c->cpuid_level >= 0x00000001) {
 		c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xFF;
-#ifdef CONFIG_X86_HT
+#ifdef CONFIG_X86_32
+# ifdef CONFIG_X86_HT
 		c->apicid = phys_pkg_id(c->initial_apicid, 0);
-		c->phys_proc_id = c->initial_apicid;
-#else
+# else
 		c->apicid = c->initial_apicid;
+# endif
+#endif
+
+#ifdef CONFIG_X86_HT
+		c->phys_proc_id = c->initial_apicid;
 #endif
 	}
 
-	if (c->extended_cpuid_level >= 0x80000004)
-		get_model_name(c); /* Default name */
+	get_model_name(c); /* Default name */
 
 	init_scattered_cpuid_features(c);
 	detect_nopl(c);
@@ -521,30 +631,29 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	c->loops_per_jiffy = loops_per_jiffy;
 	c->x86_cache_size = -1;
 	c->x86_vendor = X86_VENDOR_UNKNOWN;
-	c->cpuid_level = -1;	/* CPUID not detected */
 	c->x86_model = c->x86_mask = 0;	/* So far unknown... */
 	c->x86_vendor_id[0] = '\0'; /* Unset */
 	c->x86_model_id[0] = '\0';  /* Unset */
 	c->x86_max_cores = 1;
+	c->x86_coreid_bits = 0;
+#ifdef CONFIG_X86_64
+	c->x86_clflush_size = 64;
+#else
+	c->cpuid_level = -1;	/* CPUID not detected */
 	c->x86_clflush_size = 32;
+#endif
+	c->x86_cache_alignment = c->x86_clflush_size;
 	memset(&c->x86_capability, 0, sizeof c->x86_capability);
 
-	if (!have_cpuid_p()) {
-		/*
-		 * First of all, decide if this is a 486 or higher
-		 * It's a 486 if we can modify the AC flag
-		 */
-		if (flag_is_changeable_p(X86_EFLAGS_AC))
-			c->x86 = 4;
-		else
-			c->x86 = 3;
-	}
-
 	generic_identify(c);
 
 	if (this_cpu->c_identify)
 		this_cpu->c_identify(c);
 
+#ifdef CONFIG_X86_64
+	c->apicid = phys_pkg_id(0);
+#endif
+
 	/*
 	 * Vendor-specific initialization.  In this section we
 	 * canonicalize the feature flags, meaning if there are
@@ -578,6 +687,10 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 				c->x86, c->x86_model);
 	}
 
+#ifdef CONFIG_X86_64
+	detect_ht(c);
+#endif
+
 	/*
 	 * On SMP, boot_cpu_data holds the common feature set between
 	 * all CPUs; so make sure that we indicate which features are
@@ -594,24 +707,34 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	for (i = 0; i < NCAPINTS; i++)
 		c->x86_capability[i] &= ~cleared_cpu_caps[i];
 
+#ifdef CONFIG_X86_MCE
 	/* Init Machine Check Exception if available. */
 	mcheck_init(c);
+#endif
 
 	select_idle_routine(c);
+
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+	numa_add_cpu(smp_processor_id());
+#endif
 }
 
 void __init identify_boot_cpu(void)
 {
 	identify_cpu(&boot_cpu_data);
+#ifdef CONFIG_X86_32
 	sysenter_setup();
 	enable_sep_cpu();
+#endif
 }
 
 void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
 {
 	BUG_ON(c == &boot_cpu_data);
 	identify_cpu(c);
+#ifdef CONFIG_X86_32
 	enable_sep_cpu();
+#endif
 	mtrr_ap_init();
 }
 
@@ -709,6 +832,89 @@ __setup("clearcpuid=", setup_disablecpuid);
 
 cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
 
+#ifdef CONFIG_X86_64
+struct x8664_pda **_cpu_pda __read_mostly;
+EXPORT_SYMBOL(_cpu_pda);
+
+struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
+
+char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
+
+void __cpuinit pda_init(int cpu)
+{
+	struct x8664_pda *pda = cpu_pda(cpu);
+
+	/* Setup up data that may be needed in __get_free_pages early */
+	loadsegment(fs, 0);
+	loadsegment(gs, 0);
+	/* Memory clobbers used to order PDA accessed */
+	mb();
+	wrmsrl(MSR_GS_BASE, pda);
+	mb();
+
+	pda->cpunumber = cpu;
+	pda->irqcount = -1;
+	pda->kernelstack = (unsigned long)stack_thread_info() -
+				 PDA_STACKOFFSET + THREAD_SIZE;
+	pda->active_mm = &init_mm;
+	pda->mmu_state = 0;
+
+	if (cpu == 0) {
+		/* others are initialized in smpboot.c */
+		pda->pcurrent = &init_task;
+		pda->irqstackptr = boot_cpu_stack;
+		pda->irqstackptr += IRQSTACKSIZE - 64;
+	} else {
+		if (!pda->irqstackptr) {
+			pda->irqstackptr = (char *)
+				__get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
+			if (!pda->irqstackptr)
+				panic("cannot allocate irqstack for cpu %d",
+				      cpu);
+			pda->irqstackptr += IRQSTACKSIZE - 64;
+		}
+
+		if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE)
+			pda->nodenumber = cpu_to_node(cpu);
+	}
+}
+
+char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
+			   DEBUG_STKSZ] __page_aligned_bss;
+
+extern asmlinkage void ignore_sysret(void);
+
+/* May not be marked __init: used by software suspend */
+void syscall_init(void)
+{
+	/*
+	 * LSTAR and STAR live in a bit strange symbiosis.
+	 * They both write to the same internal register. STAR allows to
+	 * set CS/DS but only a 32bit target. LSTAR sets the 64bit rip.
+	 */
+	wrmsrl(MSR_STAR,  ((u64)__USER32_CS)<<48  | ((u64)__KERNEL_CS)<<32);
+	wrmsrl(MSR_LSTAR, system_call);
+	wrmsrl(MSR_CSTAR, ignore_sysret);
+
+#ifdef CONFIG_IA32_EMULATION
+	syscall32_cpu_init();
+#endif
+
+	/* Flags to clear on syscall */
+	wrmsrl(MSR_SYSCALL_MASK,
+	       X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
+}
+
+unsigned long kernel_eflags;
+
+/*
+ * Copies of the original ist values from the tss are only accessed during
+ * debugging, no special alignment required.
+ */
+DEFINE_PER_CPU(struct orig_ist, orig_ist);
+
+#else
+
 /* Make sure %fs is initialized properly in idle threads */
 struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs)
 {
@@ -716,13 +922,136 @@ struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs)
 	regs->fs = __KERNEL_PERCPU;
 	return regs;
 }
+#endif
 
 /*
  * cpu_init() initializes state that is per-CPU. Some data is already
  * initialized (naturally) in the bootstrap process, such as the GDT
  * and IDT. We reload them nevertheless, this function acts as a
  * 'CPU state barrier', nothing should get across.
+ * A lot of state is already set up in PDA init for 64 bit
  */
+#ifdef CONFIG_X86_64
+void __cpuinit cpu_init(void)
+{
+	int cpu = stack_smp_processor_id();
+	struct tss_struct *t = &per_cpu(init_tss, cpu);
+	struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
+	unsigned long v;
+	char *estacks = NULL;
+	struct task_struct *me;
+	int i;
+
+	/* CPU 0 is initialised in head64.c */
+	if (cpu != 0)
+		pda_init(cpu);
+	else
+		estacks = boot_exception_stacks;
+
+	me = current;
+
+	if (cpu_test_and_set(cpu, cpu_initialized))
+		panic("CPU#%d already initialized!\n", cpu);
+
+	printk(KERN_INFO "Initializing CPU#%d\n", cpu);
+
+	clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
+
+	/*
+	 * Initialize the per-CPU GDT with the boot GDT,
+	 * and set up the GDT descriptor:
+	 */
+
+	switch_to_new_gdt();
+	load_idt((const struct desc_ptr *)&idt_descr);
+
+	memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
+	syscall_init();
+
+	wrmsrl(MSR_FS_BASE, 0);
+	wrmsrl(MSR_KERNEL_GS_BASE, 0);
+	barrier();
+
+	check_efer();
+	if (cpu != 0 && x2apic)
+		enable_x2apic();
+
+	/*
+	 * set up and load the per-CPU TSS
+	 */
+	if (!orig_ist->ist[0]) {
+		static const unsigned int order[N_EXCEPTION_STACKS] = {
+		  [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
+		  [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
+		};
+		for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+			if (cpu) {
+				estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
+				if (!estacks)
+					panic("Cannot allocate exception "
+					      "stack %ld %d\n", v, cpu);
+			}
+			estacks += PAGE_SIZE << order[v];
+			orig_ist->ist[v] = t->x86_tss.ist[v] =
+					(unsigned long)estacks;
+		}
+	}
+
+	t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
+	/*
+	 * <= is required because the CPU will access up to
+	 * 8 bits beyond the end of the IO permission bitmap.
+	 */
+	for (i = 0; i <= IO_BITMAP_LONGS; i++)
+		t->io_bitmap[i] = ~0UL;
+
+	atomic_inc(&init_mm.mm_count);
+	me->active_mm = &init_mm;
+	if (me->mm)
+		BUG();
+	enter_lazy_tlb(&init_mm, me);
+
+	load_sp0(t, &current->thread);
+	set_tss_desc(cpu, t);
+	load_TR_desc();
+	load_LDT(&init_mm.context);
+
+#ifdef CONFIG_KGDB
+	/*
+	 * If the kgdb is connected no debug regs should be altered.  This
+	 * is only applicable when KGDB and a KGDB I/O module are built
+	 * into the kernel and you are using early debugging with
+	 * kgdbwait. KGDB will control the kernel HW breakpoint registers.
+	 */
+	if (kgdb_connected && arch_kgdb_ops.correct_hw_break)
+		arch_kgdb_ops.correct_hw_break();
+	else {
+#endif
+	/*
+	 * Clear all 6 debug registers:
+	 */
+
+	set_debugreg(0UL, 0);
+	set_debugreg(0UL, 1);
+	set_debugreg(0UL, 2);
+	set_debugreg(0UL, 3);
+	set_debugreg(0UL, 6);
+	set_debugreg(0UL, 7);
+#ifdef CONFIG_KGDB
+	/* If the kgdb is connected no debug regs should be altered. */
+	}
+#endif
+
+	fpu_init();
+
+	raw_local_save_flags(kernel_eflags);
+
+	if (is_uv_system())
+		uv_cpu_init();
+}
+
+#else
+
 void __cpuinit cpu_init(void)
 {
 	int cpu = smp_processor_id();
@@ -803,3 +1132,5 @@ void __cpuinit cpu_uninit(void)
 	per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm;
 }
 #endif
+
+#endif
diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c
deleted file mode 100644
index bcb48ce05d2..00000000000
--- a/arch/x86/kernel/cpu/common_64.c
+++ /dev/null
@@ -1,816 +0,0 @@
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/string.h>
-#include <linux/bootmem.h>
-#include <linux/bitops.h>
-#include <linux/module.h>
-#include <linux/kgdb.h>
-#include <linux/topology.h>
-#include <linux/delay.h>
-#include <linux/smp.h>
-#include <linux/percpu.h>
-#include <asm/i387.h>
-#include <asm/msr.h>
-#include <asm/io.h>
-#include <asm/linkage.h>
-#include <asm/mmu_context.h>
-#include <asm/mtrr.h>
-#include <asm/mce.h>
-#include <asm/pat.h>
-#include <asm/asm.h>
-#include <asm/numa.h>
-#ifdef CONFIG_X86_LOCAL_APIC
-#include <asm/mpspec.h>
-#include <asm/apic.h>
-#include <mach_apic.h>
-#endif
-#include <asm/pda.h>
-#include <asm/pgtable.h>
-#include <asm/processor.h>
-#include <asm/desc.h>
-#include <asm/atomic.h>
-#include <asm/proto.h>
-#include <asm/sections.h>
-#include <asm/setup.h>
-#include <asm/genapic.h>
-
-#include "cpu.h"
-
-static struct cpu_dev *this_cpu __cpuinitdata;
-
-/* We need valid kernel segments for data and code in long mode too
- * IRET will check the segment types  kkeil 2000/10/28
- * Also sysret mandates a special GDT layout
- */
-/* The TLS descriptors are currently at a different place compared to i386.
-   Hopefully nobody expects them at a fixed place (Wine?) */
-DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
-	[GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } },
-	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } },
-	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } },
-	[GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } },
-	[GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } },
-	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } },
-} };
-EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
-
-__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
-
-/* Current gdt points %fs at the "master" per-cpu area: after this,
- * it's on the real one. */
-void switch_to_new_gdt(void)
-{
-	struct desc_ptr gdt_descr;
-
-	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
-	gdt_descr.size = GDT_SIZE - 1;
-	load_gdt(&gdt_descr);
-}
-
-static struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
-
-static void __cpuinit default_init(struct cpuinfo_x86 *c)
-{
-	display_cacheinfo(c);
-}
-
-static struct cpu_dev __cpuinitdata default_cpu = {
-	.c_init	= default_init,
-	.c_vendor = "Unknown",
-	.c_x86_vendor = X86_VENDOR_UNKNOWN,
-};
-
-int __cpuinit get_model_name(struct cpuinfo_x86 *c)
-{
-	unsigned int *v;
-	char *p, *q;
-
-	if (c->extended_cpuid_level < 0x80000004)
-		return 0;
-
-	v = (unsigned int *) c->x86_model_id;
-	cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
-	cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
-	cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
-	c->x86_model_id[48] = 0;
-
-	/* Intel chips right-justify this string for some dumb reason;
-	   undo that brain damage */
-	p = q = &c->x86_model_id[0];
-	while (*p == ' ')
-	     p++;
-	if (p != q) {
-	     while (*p)
-		  *q++ = *p++;
-	     while (q <= &c->x86_model_id[48])
-		  *q++ = '\0';	/* Zero-pad the rest */
-	}
-
-	return 1;
-}
-
-
-void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
-{
-	unsigned int n, dummy, ebx, ecx, edx, l2size;
-
-	n = c->extended_cpuid_level;
-
-	if (n >= 0x80000005) {
-		cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
-		printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
-				edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
-		c->x86_cache_size = (ecx>>24) + (edx>>24);
-		/* On K8 L1 TLB is inclusive, so don't count it */
-		c->x86_tlbsize = 0;
-	}
-
-	if (n < 0x80000006)	/* Some chips just has a large L1. */
-		return;
-
-	cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
-	l2size = ecx >> 16;
-	c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
-
-	c->x86_cache_size = l2size;
-
-	printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
-			l2size, ecx & 0xFF);
-}
-
-void __cpuinit detect_ht(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-	u32 eax, ebx, ecx, edx;
-	int index_msb, core_bits;
-
-	if (!cpu_has(c, X86_FEATURE_HT))
-		return;
-	if (cpu_has(c, X86_FEATURE_CMP_LEGACY))
-		goto out;
-
-	if (cpu_has(c, X86_FEATURE_XTOPOLOGY))
-		return;
-
-	cpuid(1, &eax, &ebx, &ecx, &edx);
-
-	smp_num_siblings = (ebx & 0xff0000) >> 16;
-
-	if (smp_num_siblings == 1) {
-		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
-	} else if (smp_num_siblings > 1) {
-
-		if (smp_num_siblings > NR_CPUS) {
-			printk(KERN_WARNING "CPU: Unsupported number of siblings %d",
-					smp_num_siblings);
-			smp_num_siblings = 1;
-			return;
-		}
-
-		index_msb = get_count_order(smp_num_siblings);
-		c->phys_proc_id = phys_pkg_id(index_msb);
-
-		smp_num_siblings = smp_num_siblings / c->x86_max_cores;
-
-		index_msb = get_count_order(smp_num_siblings);
-
-		core_bits = get_count_order(c->x86_max_cores);
-
-		c->cpu_core_id = phys_pkg_id(index_msb) &
-					       ((1 << core_bits) - 1);
-	}
-
-out:
-	if ((c->x86_max_cores * smp_num_siblings) > 1) {
-		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
-		       c->phys_proc_id);
-		printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
-		       c->cpu_core_id);
-	}
-#endif
-}
-
-static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
-{
-	char *v = c->x86_vendor_id;
-	int i;
-	static int printed;
-
-	for (i = 0; i < X86_VENDOR_NUM; i++) {
-		if (!cpu_devs[i])
-			break;
-
-		if (!strcmp(v, cpu_devs[i]->c_ident[0]) ||
-		    (cpu_devs[i]->c_ident[1] &&
-		     !strcmp(v, cpu_devs[i]->c_ident[1]))) {
-			this_cpu = cpu_devs[i];
-			c->x86_vendor = this_cpu->c_x86_vendor;
-			return;
-		}
-	}
-
-	if (!printed) {
-		printed++;
-		printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
-		printk(KERN_ERR "CPU: Your system may be unstable.\n");
-	}
-
-	c->x86_vendor = X86_VENDOR_UNKNOWN;
-	this_cpu = &default_cpu;
-}
-
-void __cpuinit cpu_detect(struct cpuinfo_x86 *c)
-{
-	/* Get vendor name */
-	cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
-	      (unsigned int *)&c->x86_vendor_id[0],
-	      (unsigned int *)&c->x86_vendor_id[8],
-	      (unsigned int *)&c->x86_vendor_id[4]);
-
-	c->x86 = 4;
-	/* Intel-defined flags: level 0x00000001 */
-	if (c->cpuid_level >= 0x00000001) {
-		u32 junk, tfms, cap0, misc;
-		cpuid(0x00000001, &tfms, &misc, &junk, &cap0);
-		c->x86 = (tfms >> 8) & 0xf;
-		c->x86_model = (tfms >> 4) & 0xf;
-		c->x86_mask = tfms & 0xf;
-		if (c->x86 == 0xf)
-			c->x86 += (tfms >> 20) & 0xff;
-		if (c->x86 >= 0x6)
-			c->x86_model += ((tfms >> 16) & 0xf) << 4;
-		if (cap0 & (1<<19)) {
-			c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
-			c->x86_cache_alignment = c->x86_clflush_size;
-		}
-	}
-}
-
-
-static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c)
-{
-	u32 tfms, xlvl;
-	u32 ebx;
-
-	/* Intel-defined flags: level 0x00000001 */
-	if (c->cpuid_level >= 0x00000001) {
-		u32 capability, excap;
-
-		cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
-		c->x86_capability[0] = capability;
-		c->x86_capability[4] = excap;
-	}
-
-	/* AMD-defined flags: level 0x80000001 */
-	xlvl = cpuid_eax(0x80000000);
-	c->extended_cpuid_level = xlvl;
-	if ((xlvl & 0xffff0000) == 0x80000000) {
-		if (xlvl >= 0x80000001) {
-			c->x86_capability[1] = cpuid_edx(0x80000001);
-			c->x86_capability[6] = cpuid_ecx(0x80000001);
-		}
-	}
-
-	/* Transmeta-defined flags: level 0x80860001 */
-	xlvl = cpuid_eax(0x80860000);
-	if ((xlvl & 0xffff0000) == 0x80860000) {
-		/* Don't set x86_cpuid_level here for now to not confuse. */
-		if (xlvl >= 0x80860001)
-			c->x86_capability[2] = cpuid_edx(0x80860001);
-	}
-
-	if (c->extended_cpuid_level >= 0x80000007)
-		c->x86_power = cpuid_edx(0x80000007);
-
-	if (c->extended_cpuid_level >= 0x80000008) {
-		u32 eax = cpuid_eax(0x80000008);
-
-		c->x86_virt_bits = (eax >> 8) & 0xff;
-		c->x86_phys_bits = eax & 0xff;
-	}
-}
-
-/* Do some early cpuid on the boot CPU to get some parameter that are
-   needed before check_bugs. Everything advanced is in identify_cpu
-   below. */
-static void __init early_identify_cpu(struct cpuinfo_x86 *c)
-{
-
-	c->x86_clflush_size = 64;
-	c->x86_cache_alignment = c->x86_clflush_size;
-
-	memset(&c->x86_capability, 0, sizeof c->x86_capability);
-
-	c->extended_cpuid_level = 0;
-
-	cpu_detect(c);
-
-	get_cpu_vendor(c);
-
-	get_cpu_cap(c);
-
-	if (this_cpu->c_early_init)
-		this_cpu->c_early_init(c);
-
-	validate_pat_support(c);
-}
-
-void __init early_cpu_init(void)
-{
-	struct cpu_dev **cdev;
-	int count = 0;
-
-	printk("KERNEL supported cpus:\n");
-	for (cdev = __x86_cpu_dev_start; cdev < __x86_cpu_dev_end; cdev++) {
-		struct cpu_dev *cpudev = *cdev;
-		unsigned int j;
-
-		if (count >= X86_VENDOR_NUM)
-			break;
-		cpu_devs[count] = cpudev;
-		count++;
-
-		for (j = 0; j < 2; j++) {
-			if (!cpudev->c_ident[j])
-				continue;
-			printk("  %s %s\n", cpudev->c_vendor,
-				cpudev->c_ident[j]);
-		}
-	}
-
-	early_identify_cpu(&boot_cpu_data);
-}
-
-/*
- * The NOPL instruction is supposed to exist on all CPUs with
- * family >= 6, unfortunately, that's not true in practice because
- * of early VIA chips and (more importantly) broken virtualizers that
- * are not easy to detect.  Hence, probe for it based on first
- * principles.
- *
- * Note: no 64-bit chip is known to lack these, but put the code here
- * for consistency with 32 bits, and to make it utterly trivial to
- * diagnose the problem should it ever surface.
- */
-static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
-{
-	const u32 nopl_signature = 0x888c53b1; /* Random number */
-	u32 has_nopl = nopl_signature;
-
-	clear_cpu_cap(c, X86_FEATURE_NOPL);
-	if (c->x86 >= 6) {
-		asm volatile("\n"
-			     "1:      .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */
-			     "2:\n"
-			     "        .section .fixup,\"ax\"\n"
-			     "3:      xor %0,%0\n"
-			     "        jmp 2b\n"
-			     "        .previous\n"
-			     _ASM_EXTABLE(1b,3b)
-			     : "+a" (has_nopl));
-
-		if (has_nopl == nopl_signature)
-			set_cpu_cap(c, X86_FEATURE_NOPL);
-	}
-}
-
-static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
-{
-	c->extended_cpuid_level = 0;
-
-	cpu_detect(c);
-
-	get_cpu_vendor(c);
-
-	get_cpu_cap(c);
-
-	c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff;
-#ifdef CONFIG_SMP
-	c->phys_proc_id = c->initial_apicid;
-#endif
-
-	if (c->extended_cpuid_level >= 0x80000004)
-		get_model_name(c); /* Default name */
-
-	init_scattered_cpuid_features(c);
-	detect_nopl(c);
-}
-
-/*
- * This does the hard work of actually picking apart the CPU stuff...
- */
-static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
-{
-	int i;
-
-	c->loops_per_jiffy = loops_per_jiffy;
-	c->x86_cache_size = -1;
-	c->x86_vendor = X86_VENDOR_UNKNOWN;
-	c->x86_model = c->x86_mask = 0;	/* So far unknown... */
-	c->x86_vendor_id[0] = '\0'; /* Unset */
-	c->x86_model_id[0] = '\0';  /* Unset */
-	c->x86_max_cores = 1;
-	c->x86_coreid_bits = 0;
-	c->x86_clflush_size = 64;
-	c->x86_cache_alignment = c->x86_clflush_size;
-	memset(&c->x86_capability, 0, sizeof c->x86_capability);
-
-	generic_identify(c);
-
-	c->apicid = phys_pkg_id(0);
-
-	/*
-	 * Vendor-specific initialization.  In this section we
-	 * canonicalize the feature flags, meaning if there are
-	 * features a certain CPU supports which CPUID doesn't
-	 * tell us, CPUID claiming incorrect flags, or other bugs,
-	 * we handle them here.
-	 *
-	 * At the end of this section, c->x86_capability better
-	 * indicate the features this CPU genuinely supports!
-	 */
-	if (this_cpu->c_init)
-		this_cpu->c_init(c);
-
-	detect_ht(c);
-
-	/*
-	 * On SMP, boot_cpu_data holds the common feature set between
-	 * all CPUs; so make sure that we indicate which features are
-	 * common between the CPUs.  The first time this routine gets
-	 * executed, c == &boot_cpu_data.
-	 */
-	if (c != &boot_cpu_data) {
-		/* AND the already accumulated flags with these */
-		for (i = 0; i < NCAPINTS; i++)
-			boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
-	}
-
-	/* Clear all flags overriden by options */
-	for (i = 0; i < NCAPINTS; i++)
-		c->x86_capability[i] &= ~cleared_cpu_caps[i];
-
-#ifdef CONFIG_X86_MCE
-	mcheck_init(c);
-#endif
-	select_idle_routine(c);
-
-#ifdef CONFIG_NUMA
-	numa_add_cpu(smp_processor_id());
-#endif
-
-}
-
-void __init identify_boot_cpu(void)
-{
-	identify_cpu(&boot_cpu_data);
-}
-
-void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
-{
-	BUG_ON(c == &boot_cpu_data);
-	identify_cpu(c);
-	mtrr_ap_init();
-}
-
-struct msr_range {
-	unsigned min;
-	unsigned max;
-};
-
-static struct msr_range msr_range_array[] __cpuinitdata = {
-	{ 0x00000000, 0x00000418},
-	{ 0xc0000000, 0xc000040b},
-	{ 0xc0010000, 0xc0010142},
-	{ 0xc0011000, 0xc001103b},
-};
-
-static void __cpuinit print_cpu_msr(void)
-{
-	unsigned index;
-	u64 val;
-	int i;
-	unsigned index_min, index_max;
-
-	for (i = 0; i < ARRAY_SIZE(msr_range_array); i++) {
-		index_min = msr_range_array[i].min;
-		index_max = msr_range_array[i].max;
-		for (index = index_min; index < index_max; index++) {
-			if (rdmsrl_amd_safe(index, &val))
-				continue;
-			printk(KERN_INFO " MSR%08x: %016llx\n", index, val);
-		}
-	}
-}
-
-static int show_msr __cpuinitdata;
-static __init int setup_show_msr(char *arg)
-{
-	int num;
-
-	get_option(&arg, &num);
-
-	if (num > 0)
-		show_msr = num;
-	return 1;
-}
-__setup("show_msr=", setup_show_msr);
-
-static __init int setup_noclflush(char *arg)
-{
-	setup_clear_cpu_cap(X86_FEATURE_CLFLSH);
-	return 1;
-}
-__setup("noclflush", setup_noclflush);
-
-void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
-{
-	if (c->x86_model_id[0])
-		printk(KERN_CONT "%s", c->x86_model_id);
-
-	if (c->x86_mask || c->cpuid_level >= 0)
-		printk(KERN_CONT " stepping %02x\n", c->x86_mask);
-	else
-		printk(KERN_CONT "\n");
-
-#ifdef CONFIG_SMP
-	if (c->cpu_index < show_msr)
-		print_cpu_msr();
-#else
-	if (show_msr)
-		print_cpu_msr();
-#endif
-}
-
-static __init int setup_disablecpuid(char *arg)
-{
-	int bit;
-	if (get_option(&arg, &bit) && bit < NCAPINTS*32)
-		setup_clear_cpu_cap(bit);
-	else
-		return 0;
-	return 1;
-}
-__setup("clearcpuid=", setup_disablecpuid);
-
-cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
-
-struct x8664_pda **_cpu_pda __read_mostly;
-EXPORT_SYMBOL(_cpu_pda);
-
-struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
-
-char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
-
-unsigned long __supported_pte_mask __read_mostly = ~0UL;
-EXPORT_SYMBOL_GPL(__supported_pte_mask);
-
-static int do_not_nx __cpuinitdata;
-
-/* noexec=on|off
-Control non executable mappings for 64bit processes.
-
-on	Enable(default)
-off	Disable
-*/
-static int __init nonx_setup(char *str)
-{
-	if (!str)
-		return -EINVAL;
-	if (!strncmp(str, "on", 2)) {
-		__supported_pte_mask |= _PAGE_NX;
-		do_not_nx = 0;
-	} else if (!strncmp(str, "off", 3)) {
-		do_not_nx = 1;
-		__supported_pte_mask &= ~_PAGE_NX;
-	}
-	return 0;
-}
-early_param("noexec", nonx_setup);
-
-int force_personality32;
-
-/* noexec32=on|off
-Control non executable heap for 32bit processes.
-To control the stack too use noexec=off
-
-on	PROT_READ does not imply PROT_EXEC for 32bit processes (default)
-off	PROT_READ implies PROT_EXEC
-*/
-static int __init nonx32_setup(char *str)
-{
-	if (!strcmp(str, "on"))
-		force_personality32 &= ~READ_IMPLIES_EXEC;
-	else if (!strcmp(str, "off"))
-		force_personality32 |= READ_IMPLIES_EXEC;
-	return 1;
-}
-__setup("noexec32=", nonx32_setup);
-
-void pda_init(int cpu)
-{
-	struct x8664_pda *pda = cpu_pda(cpu);
-
-	/* Setup up data that may be needed in __get_free_pages early */
-	loadsegment(fs, 0);
-	loadsegment(gs, 0);
-	/* Memory clobbers used to order PDA accessed */
-	mb();
-	wrmsrl(MSR_GS_BASE, pda);
-	mb();
-
-	pda->cpunumber = cpu;
-	pda->irqcount = -1;
-	pda->kernelstack = (unsigned long)stack_thread_info() -
-				 PDA_STACKOFFSET + THREAD_SIZE;
-	pda->active_mm = &init_mm;
-	pda->mmu_state = 0;
-
-	if (cpu == 0) {
-		/* others are initialized in smpboot.c */
-		pda->pcurrent = &init_task;
-		pda->irqstackptr = boot_cpu_stack;
-		pda->irqstackptr += IRQSTACKSIZE - 64;
-	} else {
-		if (!pda->irqstackptr) {
-			pda->irqstackptr = (char *)
-				__get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
-			if (!pda->irqstackptr)
-				panic("cannot allocate irqstack for cpu %d",
-				      cpu);
-			pda->irqstackptr += IRQSTACKSIZE - 64;
-		}
-
-		if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE)
-			pda->nodenumber = cpu_to_node(cpu);
-	}
-}
-
-char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
-			   DEBUG_STKSZ] __page_aligned_bss;
-
-extern asmlinkage void ignore_sysret(void);
-
-/* May not be marked __init: used by software suspend */
-void syscall_init(void)
-{
-	/*
-	 * LSTAR and STAR live in a bit strange symbiosis.
-	 * They both write to the same internal register. STAR allows to
-	 * set CS/DS but only a 32bit target. LSTAR sets the 64bit rip.
-	 */
-	wrmsrl(MSR_STAR,  ((u64)__USER32_CS)<<48  | ((u64)__KERNEL_CS)<<32);
-	wrmsrl(MSR_LSTAR, system_call);
-	wrmsrl(MSR_CSTAR, ignore_sysret);
-
-#ifdef CONFIG_IA32_EMULATION
-	syscall32_cpu_init();
-#endif
-
-	/* Flags to clear on syscall */
-	wrmsrl(MSR_SYSCALL_MASK,
-	       X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
-}
-
-void __cpuinit check_efer(void)
-{
-	unsigned long efer;
-
-	rdmsrl(MSR_EFER, efer);
-	if (!(efer & EFER_NX) || do_not_nx)
-		__supported_pte_mask &= ~_PAGE_NX;
-}
-
-unsigned long kernel_eflags;
-
-/*
- * Copies of the original ist values from the tss are only accessed during
- * debugging, no special alignment required.
- */
-DEFINE_PER_CPU(struct orig_ist, orig_ist);
-
-/*
- * cpu_init() initializes state that is per-CPU. Some data is already
- * initialized (naturally) in the bootstrap process, such as the GDT
- * and IDT. We reload them nevertheless, this function acts as a
- * 'CPU state barrier', nothing should get across.
- * A lot of state is already set up in PDA init.
- */
-void __cpuinit cpu_init(void)
-{
-	int cpu = stack_smp_processor_id();
-	struct tss_struct *t = &per_cpu(init_tss, cpu);
-	struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
-	unsigned long v;
-	char *estacks = NULL;
-	struct task_struct *me;
-	int i;
-
-	/* CPU 0 is initialised in head64.c */
-	if (cpu != 0)
-		pda_init(cpu);
-	else
-		estacks = boot_exception_stacks;
-
-	me = current;
-
-	if (cpu_test_and_set(cpu, cpu_initialized))
-		panic("CPU#%d already initialized!\n", cpu);
-
-	printk(KERN_INFO "Initializing CPU#%d\n", cpu);
-
-	clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
-
-	/*
-	 * Initialize the per-CPU GDT with the boot GDT,
-	 * and set up the GDT descriptor:
-	 */
-
-	switch_to_new_gdt();
-	load_idt((const struct desc_ptr *)&idt_descr);
-
-	memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
-	syscall_init();
-
-	wrmsrl(MSR_FS_BASE, 0);
-	wrmsrl(MSR_KERNEL_GS_BASE, 0);
-	barrier();
-
-	check_efer();
-	if (cpu != 0 && x2apic)
-		enable_x2apic();
-
-	/*
-	 * set up and load the per-CPU TSS
-	 */
-	if (!orig_ist->ist[0]) {
-		static const unsigned int order[N_EXCEPTION_STACKS] = {
-		  [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
-		  [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
-		};
-		for (v = 0; v < N_EXCEPTION_STACKS; v++) {
-			if (cpu) {
-				estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
-				if (!estacks)
-					panic("Cannot allocate exception "
-					      "stack %ld %d\n", v, cpu);
-			}
-			estacks += PAGE_SIZE << order[v];
-			orig_ist->ist[v] = t->x86_tss.ist[v] =
-					(unsigned long)estacks;
-		}
-	}
-
-	t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
-	/*
-	 * <= is required because the CPU will access up to
-	 * 8 bits beyond the end of the IO permission bitmap.
-	 */
-	for (i = 0; i <= IO_BITMAP_LONGS; i++)
-		t->io_bitmap[i] = ~0UL;
-
-	atomic_inc(&init_mm.mm_count);
-	me->active_mm = &init_mm;
-	if (me->mm)
-		BUG();
-	enter_lazy_tlb(&init_mm, me);
-
-	load_sp0(t, &current->thread);
-	set_tss_desc(cpu, t);
-	load_TR_desc();
-	load_LDT(&init_mm.context);
-
-#ifdef CONFIG_KGDB
-	/*
-	 * If the kgdb is connected no debug regs should be altered.  This
-	 * is only applicable when KGDB and a KGDB I/O module are built
-	 * into the kernel and you are using early debugging with
-	 * kgdbwait. KGDB will control the kernel HW breakpoint registers.
-	 */
-	if (kgdb_connected && arch_kgdb_ops.correct_hw_break)
-		arch_kgdb_ops.correct_hw_break();
-	else {
-#endif
-	/*
-	 * Clear all 6 debug registers:
-	 */
-
-	set_debugreg(0UL, 0);
-	set_debugreg(0UL, 1);
-	set_debugreg(0UL, 2);
-	set_debugreg(0UL, 3);
-	set_debugreg(0UL, 6);
-	set_debugreg(0UL, 7);
-#ifdef CONFIG_KGDB
-	/* If the kgdb is connected no debug regs should be altered. */
-	}
-#endif
-
-	fpu_init();
-
-	raw_local_save_flags(kernel_eflags);
-
-	if (is_uv_system())
-		uv_cpu_init();
-}
diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h
index 3cc9d92afd8..de4094a3921 100644
--- a/arch/x86/kernel/cpu/cpu.h
+++ b/arch/x86/kernel/cpu/cpu.h
@@ -31,7 +31,6 @@ struct cpu_dev {
 
 extern struct cpu_dev *__x86_cpu_dev_start[], *__x86_cpu_dev_end[];
 
-extern int get_model_name(struct cpuinfo_x86 *c);
 extern void display_cacheinfo(struct cpuinfo_x86 *c);
 
 #endif
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index dd097b83583..c24c4a487b7 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -256,7 +256,8 @@ static u32 get_cur_val(const cpumask_t *mask)
  * Only IA32_APERF/IA32_MPERF ratio is architecturally defined and
  * no meaning should be associated with absolute values of these MSRs.
  */
-static unsigned int get_measured_perf(unsigned int cpu)
+static unsigned int get_measured_perf(struct cpufreq_policy *policy,
+				      unsigned int cpu)
 {
 	union {
 		struct {
@@ -326,7 +327,7 @@ static unsigned int get_measured_perf(unsigned int cpu)
 
 #endif
 
-	retval = per_cpu(drv_data, cpu)->max_freq * perf_percent / 100;
+	retval = per_cpu(drv_data, policy->cpu)->max_freq * perf_percent / 100;
 
 	put_cpu();
 	set_cpus_allowed_ptr(current, &saved_mask);
@@ -785,7 +786,11 @@ static int __init acpi_cpufreq_init(void)
 	if (ret)
 		return ret;
 
-	return cpufreq_register_driver(&acpi_cpufreq_driver);
+	ret = cpufreq_register_driver(&acpi_cpufreq_driver);
+	if (ret)
+		free_percpu(acpi_perf_data);
+
+	return ret;
 }
 
 static void __exit acpi_cpufreq_exit(void)
@@ -795,8 +800,6 @@ static void __exit acpi_cpufreq_exit(void)
 	cpufreq_unregister_driver(&acpi_cpufreq_driver);
 
 	free_percpu(acpi_perf_data);
-
-	return;
 }
 
 module_param(acpi_pstate_strict, uint, 0644);
diff --git a/arch/x86/kernel/cpu/cpufreq/elanfreq.c b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
index e4a4bf870e9..fe613c93b36 100644
--- a/arch/x86/kernel/cpu/cpufreq/elanfreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
@@ -25,8 +25,8 @@
 #include <linux/cpufreq.h>
 
 #include <asm/msr.h>
-#include <asm/timex.h>
-#include <asm/io.h>
+#include <linux/timex.h>
+#include <linux/io.h>
 
 #define REG_CSCIR 0x22		/* Chip Setup and Control Index Register    */
 #define REG_CSCDR 0x23		/* Chip Setup and Control Data  Register    */
@@ -82,7 +82,7 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
 	u8 clockspeed_reg;    /* Clock Speed Register */
 
 	local_irq_disable();
-	outb_p(0x80,REG_CSCIR);
+	outb_p(0x80, REG_CSCIR);
 	clockspeed_reg = inb_p(REG_CSCDR);
 	local_irq_enable();
 
@@ -98,10 +98,10 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
 	}
 
 	/* 33 MHz is not 32 MHz... */
-	if ((clockspeed_reg & 0xE0)==0xA0)
+	if ((clockspeed_reg & 0xE0) == 0xA0)
 		return 33000;
 
-	return ((1<<((clockspeed_reg & 0xE0) >> 5)) * 1000);
+	return (1<<((clockspeed_reg & 0xE0) >> 5)) * 1000;
 }
 
 
@@ -117,7 +117,7 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
  *	There is no return value.
  */
 
-static void elanfreq_set_cpu_state (unsigned int state)
+static void elanfreq_set_cpu_state(unsigned int state)
 {
 	struct cpufreq_freqs    freqs;
 
@@ -144,20 +144,20 @@ static void elanfreq_set_cpu_state (unsigned int state)
 	 */
 
 	local_irq_disable();
-	outb_p(0x40,REG_CSCIR);		/* Disable hyperspeed mode */
-	outb_p(0x00,REG_CSCDR);
+	outb_p(0x40, REG_CSCIR);		/* Disable hyperspeed mode */
+	outb_p(0x00, REG_CSCDR);
 	local_irq_enable();		/* wait till internal pipelines and */
 	udelay(1000);			/* buffers have cleaned up          */
 
 	local_irq_disable();
 
 	/* now, set the CPU clock speed register (0x80) */
-	outb_p(0x80,REG_CSCIR);
-	outb_p(elan_multiplier[state].val80h,REG_CSCDR);
+	outb_p(0x80, REG_CSCIR);
+	outb_p(elan_multiplier[state].val80h, REG_CSCDR);
 
 	/* now, the hyperspeed bit in PMU Force Mode Register (0x40) */
-	outb_p(0x40,REG_CSCIR);
-	outb_p(elan_multiplier[state].val40h,REG_CSCDR);
+	outb_p(0x40, REG_CSCIR);
+	outb_p(elan_multiplier[state].val40h, REG_CSCDR);
 	udelay(10000);
 	local_irq_enable();
 
@@ -173,12 +173,12 @@ static void elanfreq_set_cpu_state (unsigned int state)
  *	for the hardware supported by the driver.
  */
 
-static int elanfreq_verify (struct cpufreq_policy *policy)
+static int elanfreq_verify(struct cpufreq_policy *policy)
 {
 	return cpufreq_frequency_table_verify(policy, &elanfreq_table[0]);
 }
 
-static int elanfreq_target (struct cpufreq_policy *policy,
+static int elanfreq_target(struct cpufreq_policy *policy,
 			    unsigned int target_freq,
 			    unsigned int relation)
 {
@@ -205,7 +205,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 
 	/* capability check */
 	if ((c->x86_vendor != X86_VENDOR_AMD) ||
-	    (c->x86 != 4) || (c->x86_model!=10))
+	    (c->x86 != 4) || (c->x86_model != 10))
 		return -ENODEV;
 
 	/* max freq */
@@ -213,7 +213,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 		max_freq = elanfreq_get_cpu_frequency(0);
 
 	/* table init */
-	for (i=0; (elanfreq_table[i].frequency != CPUFREQ_TABLE_END); i++) {
+	for (i = 0; (elanfreq_table[i].frequency != CPUFREQ_TABLE_END); i++) {
 		if (elanfreq_table[i].frequency > max_freq)
 			elanfreq_table[i].frequency = CPUFREQ_ENTRY_INVALID;
 	}
@@ -224,7 +224,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 
 	result = cpufreq_frequency_table_cpuinfo(policy, elanfreq_table);
 	if (result)
-		return (result);
+		return result;
 
 	cpufreq_frequency_table_get_attr(elanfreq_table, policy->cpu);
 	return 0;
@@ -260,7 +260,7 @@ __setup("elanfreq=", elanfreq_setup);
 #endif
 
 
-static struct freq_attr* elanfreq_attr[] = {
+static struct freq_attr *elanfreq_attr[] = {
 	&cpufreq_freq_attr_scaling_available_freqs,
 	NULL,
 };
@@ -284,9 +284,9 @@ static int __init elanfreq_init(void)
 
 	/* Test if we have the right hardware */
 	if ((c->x86_vendor != X86_VENDOR_AMD) ||
-		(c->x86 != 4) || (c->x86_model!=10)) {
+		(c->x86 != 4) || (c->x86_model != 10)) {
 		printk(KERN_INFO "elanfreq: error: no Elan processor found!\n");
-                return -ENODEV;
+		return -ENODEV;
 	}
 	return cpufreq_register_driver(&elanfreq_driver);
 }
@@ -298,7 +298,7 @@ static void __exit elanfreq_exit(void)
 }
 
 
-module_param (max_freq, int, 0444);
+module_param(max_freq, int, 0444);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Robert Schwebel <r.schwebel@pengutronix.de>, Sven Geggus <sven@geggus.net>");
diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
index f1685fb91fb..b8e05ee4f73 100644
--- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
+++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
@@ -171,7 +171,7 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c)
 	}
 
 	if (c->x86 != 0xF) {
-		printk(KERN_WARNING PFX "Unknown p4-clockmod-capable CPU. Please send an e-mail to <cpufreq@lists.linux.org.uk>\n");
+		printk(KERN_WARNING PFX "Unknown p4-clockmod-capable CPU. Please send an e-mail to <cpufreq@vger.kernel.org>\n");
 		return 0;
 	}
 
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
index eb9b62b0830..b5ced806a31 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
@@ -15,12 +15,11 @@
 #include <linux/slab.h>
 
 #include <asm/msr.h>
-#include <asm/timex.h>
-#include <asm/io.h>
+#include <linux/timex.h>
+#include <linux/io.h>
 
-
-#define POWERNOW_IOPORT 0xfff0         /* it doesn't matter where, as long
-					  as it is unused */
+#define POWERNOW_IOPORT 0xfff0          /* it doesn't matter where, as long
+					   as it is unused */
 
 static unsigned int                     busfreq;   /* FSB, in 10 kHz */
 static unsigned int                     max_multiplier;
@@ -53,7 +52,7 @@ static int powernow_k6_get_cpu_multiplier(void)
 
 	msrval = POWERNOW_IOPORT + 0x1;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */
-	invalue=inl(POWERNOW_IOPORT + 0x8);
+	invalue = inl(POWERNOW_IOPORT + 0x8);
 	msrval = POWERNOW_IOPORT + 0x0;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */
 
@@ -67,9 +66,9 @@ static int powernow_k6_get_cpu_multiplier(void)
  *
  *   Tries to change the PowerNow! multiplier
  */
-static void powernow_k6_set_state (unsigned int best_i)
+static void powernow_k6_set_state(unsigned int best_i)
 {
-	unsigned long           outvalue=0, invalue=0;
+	unsigned long           outvalue = 0, invalue = 0;
 	unsigned long           msrval;
 	struct cpufreq_freqs    freqs;
 
@@ -90,10 +89,10 @@ static void powernow_k6_set_state (unsigned int best_i)
 
 	msrval = POWERNOW_IOPORT + 0x1;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */
-	invalue=inl(POWERNOW_IOPORT + 0x8);
+	invalue = inl(POWERNOW_IOPORT + 0x8);
 	invalue = invalue & 0xf;
 	outvalue = outvalue | invalue;
-	outl(outvalue ,(POWERNOW_IOPORT + 0x8));
+	outl(outvalue , (POWERNOW_IOPORT + 0x8));
 	msrval = POWERNOW_IOPORT + 0x0;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */
 
@@ -124,7 +123,7 @@ static int powernow_k6_verify(struct cpufreq_policy *policy)
  *
  * sets a new CPUFreq policy
  */
-static int powernow_k6_target (struct cpufreq_policy *policy,
+static int powernow_k6_target(struct cpufreq_policy *policy,
 			       unsigned int target_freq,
 			       unsigned int relation)
 {
@@ -152,7 +151,7 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 	busfreq = cpu_khz / max_multiplier;
 
 	/* table init */
-	for (i=0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) {
+	for (i = 0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) {
 		if (clock_ratio[i].index > max_multiplier)
 			clock_ratio[i].frequency = CPUFREQ_ENTRY_INVALID;
 		else
@@ -165,7 +164,7 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 
 	result = cpufreq_frequency_table_cpuinfo(policy, clock_ratio);
 	if (result)
-		return (result);
+		return result;
 
 	cpufreq_frequency_table_get_attr(clock_ratio, policy->cpu);
 
@@ -176,8 +175,8 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 static int powernow_k6_cpu_exit(struct cpufreq_policy *policy)
 {
 	unsigned int i;
-	for (i=0; i<8; i++) {
-		if (i==max_multiplier)
+	for (i = 0; i < 8; i++) {
+		if (i == max_multiplier)
 			powernow_k6_set_state(i);
 	}
 	cpufreq_frequency_table_put_attr(policy->cpu);
@@ -189,7 +188,7 @@ static unsigned int powernow_k6_get(unsigned int cpu)
 	return busfreq * powernow_k6_get_cpu_multiplier();
 }
 
-static struct freq_attr* powernow_k6_attr[] = {
+static struct freq_attr *powernow_k6_attr[] = {
 	&cpufreq_freq_attr_scaling_available_freqs,
 	NULL,
 };
@@ -227,7 +226,7 @@ static int __init powernow_k6_init(void)
 	}
 
 	if (cpufreq_register_driver(&powernow_k6_driver)) {
-		release_region (POWERNOW_IOPORT, 16);
+		release_region(POWERNOW_IOPORT, 16);
 		return -EINVAL;
 	}
 
@@ -243,13 +242,13 @@ static int __init powernow_k6_init(void)
 static void __exit powernow_k6_exit(void)
 {
 	cpufreq_unregister_driver(&powernow_k6_driver);
-	release_region (POWERNOW_IOPORT, 16);
+	release_region(POWERNOW_IOPORT, 16);
 }
 
 
-MODULE_AUTHOR ("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>");
-MODULE_DESCRIPTION ("PowerNow! driver for AMD K6-2+ / K6-3+ processors.");
-MODULE_LICENSE ("GPL");
+MODULE_AUTHOR("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>");
+MODULE_DESCRIPTION("PowerNow! driver for AMD K6-2+ / K6-3+ processors.");
+MODULE_LICENSE("GPL");
 
 module_init(powernow_k6_init);
 module_exit(powernow_k6_exit);
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index 15e13c01cc3..3b5f06423e7 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -26,7 +26,7 @@
 #include <asm/cpufeature.h>
 
 #define PFX		"speedstep-centrino: "
-#define MAINTAINER	"cpufreq@lists.linux.org.uk"
+#define MAINTAINER	"cpufreq@vger.kernel.org"
 
 #define dprintk(msg...) \
 	cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-centrino", msg)
diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c
index 3f8c7283d81..ffd0f5ed071 100644
--- a/arch/x86/kernel/cpu/cyrix.c
+++ b/arch/x86/kernel/cpu/cyrix.c
@@ -301,7 +301,6 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 			 */
 			if ((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <= dir1 && dir1 <= 0x8f))
 				geode_configure();
-			get_model_name(c);  /* get CPU marketing name */
 			return;
 		} else { /* MediaGX */
 			Cx86_cb[2] = (dir0_lsn & 1) ? '3' : '4';
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index 959417b8cd6..99468dbd08d 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -15,6 +15,11 @@
 #include <asm/ds.h>
 #include <asm/bugs.h>
 
+#ifdef CONFIG_X86_64
+#include <asm/topology.h>
+#include <asm/numa_64.h>
+#endif
+
 #include "cpu.h"
 
 #ifdef CONFIG_X86_LOCAL_APIC
@@ -25,14 +30,20 @@
 
 static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
 {
-	/* Netburst reports 64 bytes clflush size, but does IO in 128 bytes */
-	if (c->x86 == 15 && c->x86_cache_alignment == 64)
-		c->x86_cache_alignment = 128;
 	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
 		(c->x86 == 0x6 && c->x86_model >= 0x0e))
 		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+
+#ifdef CONFIG_X86_64
+	set_cpu_cap(c, X86_FEATURE_SYSENTER32);
+#else
+	/* Netburst reports 64 bytes clflush size, but does IO in 128 bytes */
+	if (c->x86 == 15 && c->x86_cache_alignment == 64)
+		c->x86_cache_alignment = 128;
+#endif
 }
 
+#ifdef CONFIG_X86_32
 /*
  *	Early probe support logic for ppro memory erratum #50
  *
@@ -52,15 +63,54 @@ int __cpuinit ppro_with_ram_bug(void)
 	return 0;
 }
 
+#ifdef CONFIG_X86_F00F_BUG
+static void __cpuinit trap_init_f00f_bug(void)
+{
+	__set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
 
-/*
- * P4 Xeon errata 037 workaround.
- * Hardware prefetcher may cause stale data to be loaded into the cache.
- */
-static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
+	/*
+	 * Update the IDT descriptor and reload the IDT so that
+	 * it uses the read-only mapped virtual address.
+	 */
+	idt_descr.address = fix_to_virt(FIX_F00F_IDT);
+	load_idt(&idt_descr);
+}
+#endif
+
+static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
 {
 	unsigned long lo, hi;
 
+#ifdef CONFIG_X86_F00F_BUG
+	/*
+	 * All current models of Pentium and Pentium with MMX technology CPUs
+	 * have the F0 0F bug, which lets nonprivileged users lock up the system.
+	 * Note that the workaround only should be initialized once...
+	 */
+	c->f00f_bug = 0;
+	if (!paravirt_enabled() && c->x86 == 5) {
+		static int f00f_workaround_enabled;
+
+		c->f00f_bug = 1;
+		if (!f00f_workaround_enabled) {
+			trap_init_f00f_bug();
+			printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n");
+			f00f_workaround_enabled = 1;
+		}
+	}
+#endif
+
+	/*
+	 * SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until
+	 * model 3 mask 3
+	 */
+	if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633)
+		clear_cpu_cap(c, X86_FEATURE_SEP);
+
+	/*
+	 * P4 Xeon errata 037 workaround.
+	 * Hardware prefetcher may cause stale data to be loaded into the cache.
+	 */
 	if ((c->x86 == 15) && (c->x86_model == 1) && (c->x86_mask == 1)) {
 		rdmsr(MSR_IA32_MISC_ENABLE, lo, hi);
 		if ((lo & (1<<9)) == 0) {
@@ -70,13 +120,68 @@ static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
 			wrmsr (MSR_IA32_MISC_ENABLE, lo, hi);
 		}
 	}
+
+	/*
+	 * See if we have a good local APIC by checking for buggy Pentia,
+	 * i.e. all B steppings and the C2 stepping of P54C when using their
+	 * integrated APIC (see 11AP erratum in "Pentium Processor
+	 * Specification Update").
+	 */
+	if (cpu_has_apic && (c->x86<<8 | c->x86_model<<4) == 0x520 &&
+	    (c->x86_mask < 0x6 || c->x86_mask == 0xb))
+		set_cpu_cap(c, X86_FEATURE_11AP);
+
+
+#ifdef CONFIG_X86_INTEL_USERCOPY
+	/*
+	 * Set up the preferred alignment for movsl bulk memory moves
+	 */
+	switch (c->x86) {
+	case 4:		/* 486: untested */
+		break;
+	case 5:		/* Old Pentia: untested */
+		break;
+	case 6:		/* PII/PIII only like movsl with 8-byte alignment */
+		movsl_mask.mask = 7;
+		break;
+	case 15:	/* P4 is OK down to 8-byte alignment */
+		movsl_mask.mask = 7;
+		break;
+	}
+#endif
+
+#ifdef CONFIG_X86_NUMAQ
+	numaq_tsc_disable();
+#endif
+}
+#else
+static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
+{
 }
+#endif
 
+static void __cpuinit srat_detect_node(void)
+{
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+	unsigned node;
+	int cpu = smp_processor_id();
+	int apicid = hard_smp_processor_id();
+
+	/* Don't do the funky fallback heuristics the AMD version employs
+	   for now. */
+	node = apicid_to_node[apicid];
+	if (node == NUMA_NO_NODE || !node_online(node))
+		node = first_node(node_online_map);
+	numa_set_node(cpu, node);
+
+	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
+#endif
+}
 
 /*
  * find out the number of processor cores on the die
  */
-static int __cpuinit num_cpu_cores(struct cpuinfo_x86 *c)
+static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
 {
 	unsigned int eax, ebx, ecx, edx;
 
@@ -91,45 +196,51 @@ static int __cpuinit num_cpu_cores(struct cpuinfo_x86 *c)
 		return 1;
 }
 
-#ifdef CONFIG_X86_F00F_BUG
-static void __cpuinit trap_init_f00f_bug(void)
+static void __cpuinit detect_vmx_virtcap(struct cpuinfo_x86 *c)
 {
-	__set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
-
-	/*
-	 * Update the IDT descriptor and reload the IDT so that
-	 * it uses the read-only mapped virtual address.
-	 */
-	idt_descr.address = fix_to_virt(FIX_F00F_IDT);
-	load_idt(&idt_descr);
+	/* Intel VMX MSR indicated features */
+#define X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW	0x00200000
+#define X86_VMX_FEATURE_PROC_CTLS_VNMI		0x00400000
+#define X86_VMX_FEATURE_PROC_CTLS_2ND_CTLS	0x80000000
+#define X86_VMX_FEATURE_PROC_CTLS2_VIRT_APIC	0x00000001
+#define X86_VMX_FEATURE_PROC_CTLS2_EPT		0x00000002
+#define X86_VMX_FEATURE_PROC_CTLS2_VPID		0x00000020
+
+	u32 vmx_msr_low, vmx_msr_high, msr_ctl, msr_ctl2;
+
+	clear_cpu_cap(c, X86_FEATURE_TPR_SHADOW);
+	clear_cpu_cap(c, X86_FEATURE_VNMI);
+	clear_cpu_cap(c, X86_FEATURE_FLEXPRIORITY);
+	clear_cpu_cap(c, X86_FEATURE_EPT);
+	clear_cpu_cap(c, X86_FEATURE_VPID);
+
+	rdmsr(MSR_IA32_VMX_PROCBASED_CTLS, vmx_msr_low, vmx_msr_high);
+	msr_ctl = vmx_msr_high | vmx_msr_low;
+	if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW)
+		set_cpu_cap(c, X86_FEATURE_TPR_SHADOW);
+	if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_VNMI)
+		set_cpu_cap(c, X86_FEATURE_VNMI);
+	if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_2ND_CTLS) {
+		rdmsr(MSR_IA32_VMX_PROCBASED_CTLS2,
+		      vmx_msr_low, vmx_msr_high);
+		msr_ctl2 = vmx_msr_high | vmx_msr_low;
+		if ((msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_VIRT_APIC) &&
+		    (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW))
+			set_cpu_cap(c, X86_FEATURE_FLEXPRIORITY);
+		if (msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_EPT)
+			set_cpu_cap(c, X86_FEATURE_EPT);
+		if (msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_VPID)
+			set_cpu_cap(c, X86_FEATURE_VPID);
+	}
 }
-#endif
 
 static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 {
 	unsigned int l2 = 0;
-	char *p = NULL;
 
 	early_init_intel(c);
 
-#ifdef CONFIG_X86_F00F_BUG
-	/*
-	 * All current models of Pentium and Pentium with MMX technology CPUs
-	 * have the F0 0F bug, which lets nonprivileged users lock up the system.
-	 * Note that the workaround only should be initialized once...
-	 */
-	c->f00f_bug = 0;
-	if (!paravirt_enabled() && c->x86 == 5) {
-		static int f00f_workaround_enabled;
-
-		c->f00f_bug = 1;
-		if (!f00f_workaround_enabled) {
-			trap_init_f00f_bug();
-			printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n");
-			f00f_workaround_enabled = 1;
-		}
-	}
-#endif
+	intel_workarounds(c);
 
 	l2 = init_intel_cacheinfo(c);
 	if (c->cpuid_level > 9) {
@@ -139,16 +250,32 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 			set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON);
 	}
 
-	/* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */
-	if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633)
-		clear_cpu_cap(c, X86_FEATURE_SEP);
+	if (cpu_has_xmm2)
+		set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
+	if (cpu_has_ds) {
+		unsigned int l1;
+		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
+		if (!(l1 & (1<<11)))
+			set_cpu_cap(c, X86_FEATURE_BTS);
+		if (!(l1 & (1<<12)))
+			set_cpu_cap(c, X86_FEATURE_PEBS);
+		ds_init_intel(c);
+	}
 
+#ifdef CONFIG_X86_64
+	if (c->x86 == 15)
+		c->x86_cache_alignment = c->x86_clflush_size * 2;
+	if (c->x86 == 6)
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+#else
 	/*
 	 * Names for the Pentium II/Celeron processors
 	 * detectable only by also checking the cache size.
 	 * Dixon is NOT a Celeron.
 	 */
 	if (c->x86 == 6) {
+		char *p = NULL;
+
 		switch (c->x86_model) {
 		case 5:
 			if (c->x86_mask == 0) {
@@ -171,77 +298,41 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 				p = "Celeron (Coppermine)";
 			break;
 		}
+
+		if (p)
+			strcpy(c->x86_model_id, p);
 	}
 
-	if (p)
-		strcpy(c->x86_model_id, p);
+	if (c->x86 == 15)
+		set_cpu_cap(c, X86_FEATURE_P4);
+	if (c->x86 == 6)
+		set_cpu_cap(c, X86_FEATURE_P3);
 
-	detect_extended_topology(c);
+	if (cpu_has_bts)
+		ptrace_bts_init_intel(c);
 
+#endif
+
+	detect_extended_topology(c);
 	if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) {
 		/*
 		 * let's use the legacy cpuid vector 0x1 and 0x4 for topology
 		 * detection.
 		 */
-		c->x86_max_cores = num_cpu_cores(c);
+		c->x86_max_cores = intel_num_cpu_cores(c);
+#ifdef CONFIG_X86_32
 		detect_ht(c);
-	}
-
-	/* Work around errata */
-	Intel_errata_workarounds(c);
-
-#ifdef CONFIG_X86_INTEL_USERCOPY
-	/*
-	 * Set up the preferred alignment for movsl bulk memory moves
-	 */
-	switch (c->x86) {
-	case 4:		/* 486: untested */
-		break;
-	case 5:		/* Old Pentia: untested */
-		break;
-	case 6:		/* PII/PIII only like movsl with 8-byte alignment */
-		movsl_mask.mask = 7;
-		break;
-	case 15:	/* P4 is OK down to 8-byte alignment */
-		movsl_mask.mask = 7;
-		break;
-	}
 #endif
-
-	if (cpu_has_xmm2)
-		set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
-	if (c->x86 == 15) {
-		set_cpu_cap(c, X86_FEATURE_P4);
 	}
-	if (c->x86 == 6)
-		set_cpu_cap(c, X86_FEATURE_P3);
-	if (cpu_has_ds) {
-		unsigned int l1;
-		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
-		if (!(l1 & (1<<11)))
-			set_cpu_cap(c, X86_FEATURE_BTS);
-		if (!(l1 & (1<<12)))
-			set_cpu_cap(c, X86_FEATURE_PEBS);
-	}
-
-	if (cpu_has_bts)
-		ds_init_intel(c);
 
-	/*
-	 * See if we have a good local APIC by checking for buggy Pentia,
-	 * i.e. all B steppings and the C2 stepping of P54C when using their
-	 * integrated APIC (see 11AP erratum in "Pentium Processor
-	 * Specification Update").
-	 */
-	if (cpu_has_apic && (c->x86<<8 | c->x86_model<<4) == 0x520 &&
-	    (c->x86_mask < 0x6 || c->x86_mask == 0xb))
-		set_cpu_cap(c, X86_FEATURE_11AP);
+	/* Work around errata */
+	srat_detect_node();
 
-#ifdef CONFIG_X86_NUMAQ
-	numaq_tsc_disable();
-#endif
+	if (cpu_has(c, X86_FEATURE_VMX))
+		detect_vmx_virtcap(c);
 }
 
+#ifdef CONFIG_X86_32
 static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned int size)
 {
 	/*
@@ -254,10 +345,12 @@ static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned i
 		size = 256;
 	return size;
 }
+#endif
 
 static struct cpu_dev intel_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Intel",
 	.c_ident	= { "GenuineIntel" },
+#ifdef CONFIG_X86_32
 	.c_models = {
 		{ .vendor = X86_VENDOR_INTEL, .family = 4, .model_names =
 		  {
@@ -307,13 +400,12 @@ static struct cpu_dev intel_cpu_dev __cpuinitdata = {
 		  }
 		},
 	},
+	.c_size_cache	= intel_size_cache,
+#endif
 	.c_early_init   = early_init_intel,
 	.c_init		= init_intel,
-	.c_size_cache	= intel_size_cache,
 	.c_x86_vendor	= X86_VENDOR_INTEL,
 };
 
 cpu_dev_register(intel_cpu_dev);
 
-/* arch_initcall(intel_cpu_init); */
-
diff --git a/arch/x86/kernel/cpu/intel_64.c b/arch/x86/kernel/cpu/intel_64.c
deleted file mode 100644
index 0c0a58dfe09..00000000000
--- a/arch/x86/kernel/cpu/intel_64.c
+++ /dev/null
@@ -1,99 +0,0 @@
-#include <linux/init.h>
-#include <linux/smp.h>
-#include <asm/processor.h>
-#include <asm/ptrace.h>
-#include <asm/topology.h>
-#include <asm/numa_64.h>
-
-#include "cpu.h"
-
-static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
-{
-	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
-	    (c->x86 == 0x6 && c->x86_model >= 0x0e))
-		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-
-	set_cpu_cap(c, X86_FEATURE_SYSENTER32);
-}
-
-/*
- * find out the number of processor cores on the die
- */
-static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
-{
-	unsigned int eax, t;
-
-	if (c->cpuid_level < 4)
-		return 1;
-
-	cpuid_count(4, 0, &eax, &t, &t, &t);
-
-	if (eax & 0x1f)
-		return ((eax >> 26) + 1);
-	else
-		return 1;
-}
-
-static void __cpuinit srat_detect_node(void)
-{
-#ifdef CONFIG_NUMA
-	unsigned node;
-	int cpu = smp_processor_id();
-	int apicid = hard_smp_processor_id();
-
-	/* Don't do the funky fallback heuristics the AMD version employs
-	   for now. */
-	node = apicid_to_node[apicid];
-	if (node == NUMA_NO_NODE || !node_online(node))
-		node = first_node(node_online_map);
-	numa_set_node(cpu, node);
-
-	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
-#endif
-}
-
-static void __cpuinit init_intel(struct cpuinfo_x86 *c)
-{
-	init_intel_cacheinfo(c);
-	if (c->cpuid_level > 9) {
-		unsigned eax = cpuid_eax(10);
-		/* Check for version and the number of counters */
-		if ((eax & 0xff) && (((eax>>8) & 0xff) > 1))
-			set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON);
-	}
-
-	if (cpu_has_ds) {
-		unsigned int l1, l2;
-		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
-		if (!(l1 & (1<<11)))
-			set_cpu_cap(c, X86_FEATURE_BTS);
-		if (!(l1 & (1<<12)))
-			set_cpu_cap(c, X86_FEATURE_PEBS);
-	}
-
-
-	if (cpu_has_bts)
-		ds_init_intel(c);
-
-	if (c->x86 == 15)
-		c->x86_cache_alignment = c->x86_clflush_size * 2;
-	if (c->x86 == 6)
-		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-	set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
-
-	detect_extended_topology(c);
-	if (!cpu_has(c, X86_FEATURE_XTOPOLOGY))
-		c->x86_max_cores = intel_num_cpu_cores(c);
-
-	srat_detect_node();
-}
-
-static struct cpu_dev intel_cpu_dev __cpuinitdata = {
-	.c_vendor	= "Intel",
-	.c_ident	= { "GenuineIntel" },
-	.c_early_init   = early_init_intel,
-	.c_init		= init_intel,
-	.c_x86_vendor	= X86_VENDOR_INTEL,
-};
-
-cpu_dev_register(intel_cpu_dev);
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 726a5fcdf34..4b031a4ac85 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -860,7 +860,7 @@ error:
 	return err;
 }
 
-static void mce_remove_device(unsigned int cpu)
+static __cpuinit void mce_remove_device(unsigned int cpu)
 {
 	int i;
 
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index cb7d3b6a80e..4e8d77f01ee 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -401,12 +401,7 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base,
 		tmp |= ~((1<<(hi - 1)) - 1);
 
 		if (tmp != mask_lo) {
-			static int once = 1;
-
-			if (once) {
-				printk(KERN_INFO "mtrr: your BIOS has set up an incorrect mask, fixing it up.\n");
-				once = 0;
-			}
+			WARN_ONCE(1, KERN_INFO "mtrr: your BIOS has set up an incorrect mask, fixing it up.\n");
 			mask_lo = tmp;
 		}
 	}
diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c
index 84c480bb371..4c4214690dd 100644
--- a/arch/x86/kernel/cpu/mtrr/if.c
+++ b/arch/x86/kernel/cpu/mtrr/if.c
@@ -405,9 +405,9 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset)
 			}
 			/* RED-PEN: base can be > 32bit */ 
 			len += seq_printf(seq, 
-				   "reg%02i: base=0x%05lx000 (%4luMB), size=%4lu%cB: %s, count=%d\n",
+				   "reg%02i: base=0x%06lx000 (%5luMB), size=%5lu%cB, count=%d: %s\n",
 			     i, base, base >> (20 - PAGE_SHIFT), size, factor,
-			     mtrr_attrib_to_str(type), mtrr_usage_table[i]);
+			     mtrr_usage_table[i], mtrr_attrib_to_str(type));
 		}
 	}
 	return 0;
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 58ac5d3d436..c78c04821ea 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -759,7 +759,8 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
 	/* take out UC ranges */
 	for (i = 0; i < num_var_ranges; i++) {
 		type = range_state[i].type;
-		if (type != MTRR_TYPE_UNCACHABLE)
+		if (type != MTRR_TYPE_UNCACHABLE &&
+		    type != MTRR_TYPE_WRPROT)
 			continue;
 		size = range_state[i].size_pfn;
 		if (!size)
@@ -834,7 +835,14 @@ static int __init enable_mtrr_cleanup_setup(char *str)
 		enable_mtrr_cleanup = 1;
 	return 0;
 }
-early_param("enble_mtrr_cleanup", enable_mtrr_cleanup_setup);
+early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup);
+
+static int __init mtrr_cleanup_debug_setup(char *str)
+{
+	debug_print = 1;
+	return 0;
+}
+early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup);
 
 struct var_mtrr_state {
 	unsigned long	range_startk;
@@ -898,6 +906,27 @@ set_var_mtrr_all(unsigned int address_bits)
 	}
 }
 
+static unsigned long to_size_factor(unsigned long sizek, char *factorp)
+{
+	char factor;
+	unsigned long base = sizek;
+
+	if (base & ((1<<10) - 1)) {
+		/* not MB alignment */
+		factor = 'K';
+	} else if (base & ((1<<20) - 1)){
+		factor = 'M';
+		base >>= 10;
+	} else {
+		factor = 'G';
+		base >>= 20;
+	}
+
+	*factorp = factor;
+
+	return base;
+}
+
 static unsigned int __init
 range_to_mtrr(unsigned int reg, unsigned long range_startk,
 	      unsigned long range_sizek, unsigned char type)
@@ -919,13 +948,21 @@ range_to_mtrr(unsigned int reg, unsigned long range_startk,
 			align = max_align;
 
 		sizek = 1 << align;
-		if (debug_print)
+		if (debug_print) {
+			char start_factor = 'K', size_factor = 'K';
+			unsigned long start_base, size_base;
+
+			start_base = to_size_factor(range_startk, &start_factor),
+			size_base = to_size_factor(sizek, &size_factor),
+
 			printk(KERN_DEBUG "Setting variable MTRR %d, "
-				"base: %ldMB, range: %ldMB, type %s\n",
-				reg, range_startk >> 10, sizek >> 10,
+				"base: %ld%cB, range: %ld%cB, type %s\n",
+				reg, start_base, start_factor,
+				size_base, size_factor,
 				(type == MTRR_TYPE_UNCACHABLE)?"UC":
 				    ((type == MTRR_TYPE_WRBACK)?"WB":"Other")
 				);
+		}
 		save_var_mtrr(reg++, range_startk, sizek, type);
 		range_startk += sizek;
 		range_sizek -= sizek;
@@ -970,6 +1007,8 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek,
 	/* try to append some small hole */
 	range0_basek = state->range_startk;
 	range0_sizek = ALIGN(state->range_sizek, chunk_sizek);
+
+	/* no increase */
 	if (range0_sizek == state->range_sizek) {
 		if (debug_print)
 			printk(KERN_DEBUG "rangeX: %016lx - %016lx\n",
@@ -980,13 +1019,40 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek,
 		return 0;
 	}
 
-	range0_sizek -= chunk_sizek;
-	if (range0_sizek && sizek) {
-	    while (range0_basek + range0_sizek > (basek + sizek)) {
-		range0_sizek -= chunk_sizek;
-		if (!range0_sizek)
-			break;
-	    }
+	/* only cut back, when it is not the last */
+	if (sizek) {
+		while (range0_basek + range0_sizek > (basek + sizek)) {
+			if (range0_sizek >= chunk_sizek)
+				range0_sizek -= chunk_sizek;
+			else
+				range0_sizek = 0;
+
+			if (!range0_sizek)
+				break;
+		}
+	}
+
+second_try:
+	range_basek = range0_basek + range0_sizek;
+
+	/* one hole in the middle */
+	if (range_basek > basek && range_basek <= (basek + sizek))
+		second_sizek = range_basek - basek;
+
+	if (range0_sizek > state->range_sizek) {
+
+		/* one hole in middle or at end */
+		hole_sizek = range0_sizek - state->range_sizek - second_sizek;
+
+		/* hole size should be less than half of range0 size */
+		if (hole_sizek >= (range0_sizek >> 1) &&
+		    range0_sizek >= chunk_sizek) {
+			range0_sizek -= chunk_sizek;
+			second_sizek = 0;
+			hole_sizek = 0;
+
+			goto second_try;
+		}
 	}
 
 	if (range0_sizek) {
@@ -996,50 +1062,28 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek,
 				(range0_basek + range0_sizek)<<10);
 		state->reg = range_to_mtrr(state->reg, range0_basek,
 				range0_sizek, MTRR_TYPE_WRBACK);
-
-	}
-
-	range_basek = range0_basek + range0_sizek;
-	range_sizek = chunk_sizek;
-
-	if (range_basek + range_sizek > basek &&
-	    range_basek + range_sizek <= (basek + sizek)) {
-		/* one hole */
-		second_basek = basek;
-		second_sizek = range_basek + range_sizek - basek;
 	}
 
-	/* if last piece, only could one hole near end */
-	if ((second_basek || !basek) &&
-	    range_sizek - (state->range_sizek - range0_sizek) - second_sizek <
-	    (chunk_sizek >> 1)) {
-		/*
-		 * one hole in middle (second_sizek is 0) or at end
-		 * (second_sizek is 0 )
-		 */
-		hole_sizek = range_sizek - (state->range_sizek - range0_sizek)
-				 - second_sizek;
-		hole_basek = range_basek + range_sizek - hole_sizek
-				 - second_sizek;
-	} else {
-		/* fallback for big hole, or several holes */
+	if (range0_sizek < state->range_sizek) {
+		/* need to handle left over */
 		range_sizek = state->range_sizek - range0_sizek;
-		second_basek = 0;
-		second_sizek = 0;
+
+		if (debug_print)
+			printk(KERN_DEBUG "range: %016lx - %016lx\n",
+				 range_basek<<10,
+				 (range_basek + range_sizek)<<10);
+		state->reg = range_to_mtrr(state->reg, range_basek,
+				 range_sizek, MTRR_TYPE_WRBACK);
 	}
 
-	if (debug_print)
-		printk(KERN_DEBUG "range: %016lx - %016lx\n", range_basek<<10,
-			 (range_basek + range_sizek)<<10);
-	state->reg = range_to_mtrr(state->reg, range_basek, range_sizek,
-					 MTRR_TYPE_WRBACK);
 	if (hole_sizek) {
+		hole_basek = range_basek - hole_sizek - second_sizek;
 		if (debug_print)
 			printk(KERN_DEBUG "hole: %016lx - %016lx\n",
-				 hole_basek<<10, (hole_basek + hole_sizek)<<10);
-		state->reg = range_to_mtrr(state->reg, hole_basek, hole_sizek,
-						 MTRR_TYPE_UNCACHABLE);
-
+				 hole_basek<<10,
+				 (hole_basek + hole_sizek)<<10);
+		state->reg = range_to_mtrr(state->reg, hole_basek,
+				 hole_sizek, MTRR_TYPE_UNCACHABLE);
 	}
 
 	return second_sizek;
@@ -1154,11 +1198,11 @@ struct mtrr_cleanup_result {
 };
 
 /*
- * gran_size: 1M, 2M, ..., 2G
- * chunk size: gran_size, ..., 4G
- * so we need (2+13)*6
+ * gran_size: 64K, 128K, 256K, 512K, 1M, 2M, ..., 2G
+ * chunk size: gran_size, ..., 2G
+ * so we need (1+16)*8
  */
-#define NUM_RESULT	90
+#define NUM_RESULT	136
 #define PSHIFT		(PAGE_SHIFT - 10)
 
 static struct mtrr_cleanup_result __initdata result[NUM_RESULT];
@@ -1168,13 +1212,14 @@ static unsigned long __initdata min_loss_pfn[RANGE_NUM];
 static int __init mtrr_cleanup(unsigned address_bits)
 {
 	unsigned long extra_remove_base, extra_remove_size;
-	unsigned long i, base, size, def, dummy;
+	unsigned long base, size, def, dummy;
 	mtrr_type type;
 	int nr_range, nr_range_new;
 	u64 chunk_size, gran_size;
 	unsigned long range_sums, range_sums_new;
 	int index_good;
 	int num_reg_good;
+	int i;
 
 	/* extra one for all 0 */
 	int num[MTRR_NUM_TYPES + 1];
@@ -1204,6 +1249,8 @@ static int __init mtrr_cleanup(unsigned address_bits)
 			continue;
 		if (!size)
 			type = MTRR_NUM_TYPES;
+		if (type == MTRR_TYPE_WRPROT)
+			type = MTRR_TYPE_UNCACHABLE;
 		num[type]++;
 	}
 
@@ -1216,23 +1263,57 @@ static int __init mtrr_cleanup(unsigned address_bits)
 		num_var_ranges - num[MTRR_NUM_TYPES])
 		return 0;
 
+	/* print original var MTRRs at first, for debugging: */
+	printk(KERN_DEBUG "original variable MTRRs\n");
+	for (i = 0; i < num_var_ranges; i++) {
+		char start_factor = 'K', size_factor = 'K';
+		unsigned long start_base, size_base;
+
+		size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10);
+		if (!size_base)
+			continue;
+
+		size_base = to_size_factor(size_base, &size_factor),
+		start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10);
+		start_base = to_size_factor(start_base, &start_factor),
+		type = range_state[i].type;
+
+		printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n",
+			i, start_base, start_factor,
+			size_base, size_factor,
+			(type == MTRR_TYPE_UNCACHABLE) ? "UC" :
+			    ((type == MTRR_TYPE_WRPROT) ? "WP" :
+			     ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other"))
+			);
+	}
+
 	memset(range, 0, sizeof(range));
 	extra_remove_size = 0;
-	if (mtrr_tom2) {
-		extra_remove_base = 1 << (32 - PAGE_SHIFT);
+	extra_remove_base = 1 << (32 - PAGE_SHIFT);
+	if (mtrr_tom2)
 		extra_remove_size =
 			(mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base;
-	}
 	nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base,
 					  extra_remove_size);
+	/*
+	 * [0, 1M) should always be coverred by var mtrr with WB
+	 * and fixed mtrrs should take effective before var mtrr for it
+	 */
+	nr_range = add_range_with_merge(range, nr_range, 0,
+					(1ULL<<(20 - PAGE_SHIFT)) - 1);
+	/* sort the ranges */
+	sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL);
+
 	range_sums = sum_ranges(range, nr_range);
 	printk(KERN_INFO "total RAM coverred: %ldM\n",
 	       range_sums >> (20 - PAGE_SHIFT));
 
 	if (mtrr_chunk_size && mtrr_gran_size) {
 		int num_reg;
+		char gran_factor, chunk_factor, lose_factor;
+		unsigned long gran_base, chunk_base, lose_base;
 
-		debug_print = 1;
+		debug_print++;
 		/* convert ranges to var ranges state */
 		num_reg = x86_setup_var_mtrrs(range, nr_range, mtrr_chunk_size,
 					      mtrr_gran_size);
@@ -1256,34 +1337,48 @@ static int __init mtrr_cleanup(unsigned address_bits)
 			result[i].lose_cover_sizek =
 				(range_sums - range_sums_new) << PSHIFT;
 
-		printk(KERN_INFO "%sgran_size: %ldM \tchunk_size: %ldM \t",
-			 result[i].bad?"*BAD*":" ", result[i].gran_sizek >> 10,
-			 result[i].chunk_sizek >> 10);
-		printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ldM \n",
+		gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
+		chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
+		lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
+		printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
+			 result[i].bad?"*BAD*":" ",
+			 gran_base, gran_factor, chunk_base, chunk_factor);
+		printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
 			 result[i].num_reg, result[i].bad?"-":"",
-			 result[i].lose_cover_sizek >> 10);
+			 lose_base, lose_factor);
 		if (!result[i].bad) {
 			set_var_mtrr_all(address_bits);
 			return 1;
 		}
 		printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, "
 		       "will find optimal one\n");
-		debug_print = 0;
+		debug_print--;
 		memset(result, 0, sizeof(result[0]));
 	}
 
 	i = 0;
 	memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn));
 	memset(result, 0, sizeof(result));
-	for (gran_size = (1ULL<<20); gran_size < (1ULL<<32); gran_size <<= 1) {
-		for (chunk_size = gran_size; chunk_size < (1ULL<<33);
+	for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) {
+		char gran_factor;
+		unsigned long gran_base;
+
+		if (debug_print)
+			gran_base = to_size_factor(gran_size >> 10, &gran_factor);
+
+		for (chunk_size = gran_size; chunk_size < (1ULL<<32);
 		     chunk_size <<= 1) {
 			int num_reg;
 
-			if (debug_print)
-				printk(KERN_INFO
-			       "\ngran_size: %lldM   chunk_size_size: %lldM\n",
-				       gran_size >> 20, chunk_size >> 20);
+			if (debug_print) {
+				char chunk_factor;
+				unsigned long chunk_base;
+
+				chunk_base = to_size_factor(chunk_size>>10, &chunk_factor),
+				printk(KERN_INFO "\n");
+				printk(KERN_INFO "gran_size: %ld%c   chunk_size: %ld%c \n",
+				       gran_base, gran_factor, chunk_base, chunk_factor);
+			}
 			if (i >= NUM_RESULT)
 				continue;
 
@@ -1326,12 +1421,18 @@ static int __init mtrr_cleanup(unsigned address_bits)
 
 	/* print out all */
 	for (i = 0; i < NUM_RESULT; i++) {
-		printk(KERN_INFO "%sgran_size: %ldM \tchunk_size: %ldM \t",
-		       result[i].bad?"*BAD* ":" ", result[i].gran_sizek >> 10,
-		       result[i].chunk_sizek >> 10);
-		printk(KERN_CONT "num_reg: %d \tlose RAM: %s%ldM\n",
-		       result[i].num_reg, result[i].bad?"-":"",
-		       result[i].lose_cover_sizek >> 10);
+		char gran_factor, chunk_factor, lose_factor;
+		unsigned long gran_base, chunk_base, lose_base;
+
+		gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
+		chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
+		lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
+		printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
+			 result[i].bad?"*BAD*":" ",
+			 gran_base, gran_factor, chunk_base, chunk_factor);
+		printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
+			 result[i].num_reg, result[i].bad?"-":"",
+			 lose_base, lose_factor);
 	}
 
 	/* try to find the optimal index */
@@ -1339,10 +1440,8 @@ static int __init mtrr_cleanup(unsigned address_bits)
 		nr_mtrr_spare_reg = num_var_ranges - 1;
 	num_reg_good = -1;
 	for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) {
-		if (!min_loss_pfn[i]) {
+		if (!min_loss_pfn[i])
 			num_reg_good = i;
-			break;
-		}
 	}
 
 	index_good = -1;
@@ -1358,21 +1457,26 @@ static int __init mtrr_cleanup(unsigned address_bits)
 	}
 
 	if (index_good != -1) {
+		char gran_factor, chunk_factor, lose_factor;
+		unsigned long gran_base, chunk_base, lose_base;
+
 		printk(KERN_INFO "Found optimal setting for mtrr clean up\n");
 		i = index_good;
-		printk(KERN_INFO "gran_size: %ldM \tchunk_size: %ldM \t",
-				result[i].gran_sizek >> 10,
-				result[i].chunk_sizek >> 10);
-		printk(KERN_CONT "num_reg: %d \tlose RAM: %ldM\n",
-				result[i].num_reg,
-				result[i].lose_cover_sizek >> 10);
+		gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
+		chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
+		lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
+		printk(KERN_INFO "gran_size: %ld%c \tchunk_size: %ld%c \t",
+			 gran_base, gran_factor, chunk_base, chunk_factor);
+		printk(KERN_CONT "num_reg: %d  \tlose RAM: %ld%c\n",
+			 result[i].num_reg, lose_base, lose_factor);
 		/* convert ranges to var ranges state */
 		chunk_size = result[i].chunk_sizek;
 		chunk_size <<= 10;
 		gran_size = result[i].gran_sizek;
 		gran_size <<= 10;
-		debug_print = 1;
+		debug_print++;
 		x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size);
+		debug_print--;
 		set_var_mtrr_all(address_bits);
 		return 1;
 	}
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index 05cc22dbd4f..6bff382094f 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -295,13 +295,19 @@ static int setup_k7_watchdog(unsigned nmi_hz)
 	/* setup the timer */
 	wrmsr(evntsel_msr, evntsel, 0);
 	write_watchdog_counter(perfctr_msr, "K7_PERFCTR0",nmi_hz);
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	evntsel |= K7_EVNTSEL_ENABLE;
-	wrmsr(evntsel_msr, evntsel, 0);
 
+	/* initialize the wd struct before enabling */
 	wd->perfctr_msr = perfctr_msr;
 	wd->evntsel_msr = evntsel_msr;
 	wd->cccr_msr = 0;  /* unused */
+
+	/* ok, everything is initialized, announce that we're set */
+	cpu_nmi_set_wd_enabled();
+
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= K7_EVNTSEL_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
 	return 1;
 }
 
@@ -379,13 +385,19 @@ static int setup_p6_watchdog(unsigned nmi_hz)
 	wrmsr(evntsel_msr, evntsel, 0);
 	nmi_hz = adjust_for_32bit_ctr(nmi_hz);
 	write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0",nmi_hz);
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	evntsel |= P6_EVNTSEL0_ENABLE;
-	wrmsr(evntsel_msr, evntsel, 0);
 
+	/* initialize the wd struct before enabling */
 	wd->perfctr_msr = perfctr_msr;
 	wd->evntsel_msr = evntsel_msr;
 	wd->cccr_msr = 0;  /* unused */
+
+	/* ok, everything is initialized, announce that we're set */
+	cpu_nmi_set_wd_enabled();
+
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= P6_EVNTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
 	return 1;
 }
 
@@ -432,6 +444,27 @@ static const struct wd_ops p6_wd_ops = {
 #define P4_CCCR_ENABLE		(1 << 12)
 #define P4_CCCR_OVF 		(1 << 31)
 
+#define P4_CONTROLS 18
+static unsigned int p4_controls[18] = {
+	MSR_P4_BPU_CCCR0,
+	MSR_P4_BPU_CCCR1,
+	MSR_P4_BPU_CCCR2,
+	MSR_P4_BPU_CCCR3,
+	MSR_P4_MS_CCCR0,
+	MSR_P4_MS_CCCR1,
+	MSR_P4_MS_CCCR2,
+	MSR_P4_MS_CCCR3,
+	MSR_P4_FLAME_CCCR0,
+	MSR_P4_FLAME_CCCR1,
+	MSR_P4_FLAME_CCCR2,
+	MSR_P4_FLAME_CCCR3,
+	MSR_P4_IQ_CCCR0,
+	MSR_P4_IQ_CCCR1,
+	MSR_P4_IQ_CCCR2,
+	MSR_P4_IQ_CCCR3,
+	MSR_P4_IQ_CCCR4,
+	MSR_P4_IQ_CCCR5,
+};
 /*
  * Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
  * CRU_ESCR0 (with any non-null event selector) through a complemented
@@ -473,6 +506,26 @@ static int setup_p4_watchdog(unsigned nmi_hz)
 		evntsel_msr = MSR_P4_CRU_ESCR0;
 		cccr_msr = MSR_P4_IQ_CCCR0;
 		cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4);
+
+		/*
+		 * If we're on the kdump kernel or other situation, we may
+		 * still have other performance counter registers set to
+		 * interrupt and they'll keep interrupting forever because
+		 * of the P4_CCCR_OVF quirk. So we need to ACK all the
+		 * pending interrupts and disable all the registers here,
+		 * before reenabling the NMI delivery. Refer to p4_rearm()
+		 * about the P4_CCCR_OVF quirk.
+		 */
+		if (reset_devices) {
+			unsigned int low, high;
+			int i;
+
+			for (i = 0; i < P4_CONTROLS; i++) {
+				rdmsr(p4_controls[i], low, high);
+				low &= ~(P4_CCCR_ENABLE | P4_CCCR_OVF);
+				wrmsr(p4_controls[i], low, high);
+			}
+		}
 	} else {
 		/* logical cpu 1 */
 		perfctr_msr = MSR_P4_IQ_PERFCTR1;
@@ -499,12 +552,17 @@ static int setup_p4_watchdog(unsigned nmi_hz)
 	wrmsr(evntsel_msr, evntsel, 0);
 	wrmsr(cccr_msr, cccr_val, 0);
 	write_watchdog_counter(perfctr_msr, "P4_IQ_COUNTER0", nmi_hz);
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	cccr_val |= P4_CCCR_ENABLE;
-	wrmsr(cccr_msr, cccr_val, 0);
+
 	wd->perfctr_msr = perfctr_msr;
 	wd->evntsel_msr = evntsel_msr;
 	wd->cccr_msr = cccr_msr;
+
+	/* ok, everything is initialized, announce that we're set */
+	cpu_nmi_set_wd_enabled();
+
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	cccr_val |= P4_CCCR_ENABLE;
+	wrmsr(cccr_msr, cccr_val, 0);
 	return 1;
 }
 
@@ -620,13 +678,17 @@ static int setup_intel_arch_watchdog(unsigned nmi_hz)
 	wrmsr(evntsel_msr, evntsel, 0);
 	nmi_hz = adjust_for_32bit_ctr(nmi_hz);
 	write_watchdog_counter32(perfctr_msr, "INTEL_ARCH_PERFCTR0", nmi_hz);
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
-	wrmsr(evntsel_msr, evntsel, 0);
 
 	wd->perfctr_msr = perfctr_msr;
 	wd->evntsel_msr = evntsel_msr;
 	wd->cccr_msr = 0;  /* unused */
+
+	/* ok, everything is initialized, announce that we're set */
+	cpu_nmi_set_wd_enabled();
+
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
 	intel_arch_wd_ops.checkbit = 1ULL << (eax.split.bit_width - 1);
 	return 1;
 }
diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c
index 7c46e6ecedc..52b3fefbd5a 100644
--- a/arch/x86/kernel/cpu/transmeta.c
+++ b/arch/x86/kernel/cpu/transmeta.c
@@ -5,6 +5,18 @@
 #include <asm/msr.h>
 #include "cpu.h"
 
+static void __cpuinit early_init_transmeta(struct cpuinfo_x86 *c)
+{
+	u32 xlvl;
+
+	/* Transmeta-defined flags: level 0x80860001 */
+	xlvl = cpuid_eax(0x80860000);
+	if ((xlvl & 0xffff0000) == 0x80860000) {
+		if (xlvl >= 0x80860001)
+			c->x86_capability[2] = cpuid_edx(0x80860001);
+	}
+}
+
 static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 {
 	unsigned int cap_mask, uk, max, dummy;
@@ -12,7 +24,8 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 	unsigned int cpu_rev, cpu_freq = 0, cpu_flags, new_cpu_rev;
 	char cpu_info[65];
 
-	get_model_name(c);	/* Same as AMD/Cyrix */
+	early_init_transmeta(c);
+
 	display_cacheinfo(c);
 
 	/* Print CMS and CPU revision */
@@ -85,23 +98,11 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 #endif
 }
 
-static void __cpuinit transmeta_identify(struct cpuinfo_x86 *c)
-{
-	u32 xlvl;
-
-	/* Transmeta-defined flags: level 0x80860001 */
-	xlvl = cpuid_eax(0x80860000);
-	if ((xlvl & 0xffff0000) == 0x80860000) {
-		if (xlvl >= 0x80860001)
-			c->x86_capability[2] = cpuid_edx(0x80860001);
-	}
-}
-
 static struct cpu_dev transmeta_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Transmeta",
 	.c_ident	= { "GenuineTMx86", "TransmetaCPU" },
+	.c_early_init	= early_init_transmeta,
 	.c_init		= init_transmeta,
-	.c_identify	= transmeta_identify,
 	.c_x86_vendor	= X86_VENDOR_TRANSMETA,
 };
 
diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c
index 8e9cd6a8ec1..6a44d646599 100644
--- a/arch/x86/kernel/cpuid.c
+++ b/arch/x86/kernel/cpuid.c
@@ -36,7 +36,6 @@
 #include <linux/smp_lock.h>
 #include <linux/major.h>
 #include <linux/fs.h>
-#include <linux/smp_lock.h>
 #include <linux/device.h>
 #include <linux/cpu.h>
 #include <linux/notifier.h>
diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c
index 15e6c6bc4a4..e90a60ef10c 100644
--- a/arch/x86/kernel/crash_dump_64.c
+++ b/arch/x86/kernel/crash_dump_64.c
@@ -7,9 +7,8 @@
 
 #include <linux/errno.h>
 #include <linux/crash_dump.h>
-
-#include <asm/uaccess.h>
-#include <asm/io.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
 
 /**
  * copy_oldmem_page - copy one page from "oldmem"
@@ -25,7 +24,7 @@
  * in the current kernel. We stitch up a pte, similar to kmap_atomic.
  */
 ssize_t copy_oldmem_page(unsigned long pfn, char *buf,
-                               size_t csize, unsigned long offset, int userbuf)
+		size_t csize, unsigned long offset, int userbuf)
 {
 	void  *vaddr;
 
@@ -33,14 +32,16 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf,
 		return 0;
 
 	vaddr = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE);
+	if (!vaddr)
+		return -ENOMEM;
 
 	if (userbuf) {
-		if (copy_to_user(buf, (vaddr + offset), csize)) {
+		if (copy_to_user(buf, vaddr + offset, csize)) {
 			iounmap(vaddr);
 			return -EFAULT;
 		}
 	} else
-	memcpy(buf, (vaddr + offset), csize);
+		memcpy(buf, vaddr + offset, csize);
 
 	iounmap(vaddr);
 	return csize;
diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
index 11c11b8ec48..2b69994fd3a 100644
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -2,26 +2,49 @@
  * Debug Store support
  *
  * This provides a low-level interface to the hardware's Debug Store
- * feature that is used for last branch recording (LBR) and
+ * feature that is used for branch trace store (BTS) and
  * precise-event based sampling (PEBS).
  *
- * Different architectures use a different DS layout/pointer size.
- * The below functions therefore work on a void*.
+ * It manages:
+ * - per-thread and per-cpu allocation of BTS and PEBS
+ * - buffer memory allocation (optional)
+ * - buffer overflow handling
+ * - buffer access
  *
+ * It assumes:
+ * - get_task_struct on all parameter tasks
+ * - current is allowed to trace parameter tasks
  *
- * Since there is no user for PEBS, yet, only LBR (or branch
- * trace store, BTS) is supported.
  *
- *
- * Copyright (C) 2007 Intel Corporation.
- * Markus Metzger <markus.t.metzger@intel.com>, Dec 2007
+ * Copyright (C) 2007-2008 Intel Corporation.
+ * Markus Metzger <markus.t.metzger@intel.com>, 2007-2008
  */
 
+
+#ifdef CONFIG_X86_DS
+
 #include <asm/ds.h>
 
 #include <linux/errno.h>
 #include <linux/string.h>
 #include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+
+
+/*
+ * The configuration for a particular DS hardware implementation.
+ */
+struct ds_configuration {
+	/* the size of the DS structure in bytes */
+	unsigned char  sizeof_ds;
+	/* the size of one pointer-typed field in the DS structure in bytes;
+	   this covers the first 8 fields related to buffer management. */
+	unsigned char  sizeof_field;
+	/* the size of a BTS/PEBS record in bytes */
+	unsigned char  sizeof_rec[2];
+};
+static struct ds_configuration ds_cfg;
 
 
 /*
@@ -44,378 +67,747 @@
  *   (interrupt occurs when write pointer passes interrupt pointer)
  * - value to which counter is reset following counter overflow
  *
- * On later architectures, the last branch recording hardware uses
- * 64bit pointers even in 32bit mode.
- *
- *
- * Branch Trace Store (BTS) records store information about control
- * flow changes. They at least provide the following information:
- * - source linear address
- * - destination linear address
+ * Later architectures use 64bit pointers throughout, whereas earlier
+ * architectures use 32bit pointers in 32bit mode.
  *
- * Netburst supported a predicated bit that had been dropped in later
- * architectures. We do not suppor it.
  *
+ * We compute the base address for the first 8 fields based on:
+ * - the field size stored in the DS configuration
+ * - the relative field position
+ * - an offset giving the start of the respective region
  *
- * In order to abstract from the actual DS and BTS layout, we describe
- * the access to the relevant fields.
- * Thanks to Andi Kleen for proposing this design.
+ * This offset is further used to index various arrays holding
+ * information for BTS and PEBS at the respective index.
  *
- * The implementation, however, is not as general as it might seem. In
- * order to stay somewhat simple and efficient, we assume an
- * underlying unsigned type (mostly a pointer type) and we expect the
- * field to be at least as big as that type.
+ * On later 32bit processors, we only access the lower 32bit of the
+ * 64bit pointer fields. The upper halves will be zeroed out.
  */
 
-/*
- * A special from_ip address to indicate that the BTS record is an
- * info record that needs to be interpreted or skipped.
- */
-#define BTS_ESCAPE_ADDRESS (-1)
+enum ds_field {
+	ds_buffer_base = 0,
+	ds_index,
+	ds_absolute_maximum,
+	ds_interrupt_threshold,
+};
 
-/*
- * A field access descriptor
- */
-struct access_desc {
-	unsigned char offset;
-	unsigned char size;
+enum ds_qualifier {
+	ds_bts  = 0,
+	ds_pebs
 };
 
+static inline unsigned long ds_get(const unsigned char *base,
+				   enum ds_qualifier qual, enum ds_field field)
+{
+	base += (ds_cfg.sizeof_field * (field + (4 * qual)));
+	return *(unsigned long *)base;
+}
+
+static inline void ds_set(unsigned char *base, enum ds_qualifier qual,
+			  enum ds_field field, unsigned long value)
+{
+	base += (ds_cfg.sizeof_field * (field + (4 * qual)));
+	(*(unsigned long *)base) = value;
+}
+
+
 /*
- * The configuration for a particular DS/BTS hardware implementation.
+ * Locking is done only for allocating BTS or PEBS resources and for
+ * guarding context and buffer memory allocation.
+ *
+ * Most functions require the current task to own the ds context part
+ * they are going to access. All the locking is done when validating
+ * access to the context.
  */
-struct ds_configuration {
-	/* the DS configuration */
-	unsigned char  sizeof_ds;
-	struct access_desc bts_buffer_base;
-	struct access_desc bts_index;
-	struct access_desc bts_absolute_maximum;
-	struct access_desc bts_interrupt_threshold;
-	/* the BTS configuration */
-	unsigned char  sizeof_bts;
-	struct access_desc from_ip;
-	struct access_desc to_ip;
-	/* BTS variants used to store additional information like
-	   timestamps */
-	struct access_desc info_type;
-	struct access_desc info_data;
-	unsigned long debugctl_mask;
-};
+static spinlock_t ds_lock = __SPIN_LOCK_UNLOCKED(ds_lock);
 
 /*
- * The global configuration used by the below accessor functions
+ * Validate that the current task is allowed to access the BTS/PEBS
+ * buffer of the parameter task.
+ *
+ * Returns 0, if access is granted; -Eerrno, otherwise.
  */
-static struct ds_configuration ds_cfg;
+static inline int ds_validate_access(struct ds_context *context,
+				     enum ds_qualifier qual)
+{
+	if (!context)
+		return -EPERM;
+
+	if (context->owner[qual] == current)
+		return 0;
+
+	return -EPERM;
+}
+
 
 /*
- * Accessor functions for some DS and BTS fields using the above
- * global ptrace_bts_cfg.
+ * We either support (system-wide) per-cpu or per-thread allocation.
+ * We distinguish the two based on the task_struct pointer, where a
+ * NULL pointer indicates per-cpu allocation for the current cpu.
+ *
+ * Allocations are use-counted. As soon as resources are allocated,
+ * further allocations must be of the same type (per-cpu or
+ * per-thread). We model this by counting allocations (i.e. the number
+ * of tracers of a certain type) for one type negatively:
+ *   =0  no tracers
+ *   >0  number of per-thread tracers
+ *   <0  number of per-cpu tracers
+ *
+ * The below functions to get and put tracers and to check the
+ * allocation type require the ds_lock to be held by the caller.
+ *
+ * Tracers essentially gives the number of ds contexts for a certain
+ * type of allocation.
  */
-static inline unsigned long get_bts_buffer_base(char *base)
+static long tracers;
+
+static inline void get_tracer(struct task_struct *task)
 {
-	return *(unsigned long *)(base + ds_cfg.bts_buffer_base.offset);
+	tracers += (task ? 1 : -1);
 }
-static inline void set_bts_buffer_base(char *base, unsigned long value)
+
+static inline void put_tracer(struct task_struct *task)
 {
-	(*(unsigned long *)(base + ds_cfg.bts_buffer_base.offset)) = value;
+	tracers -= (task ? 1 : -1);
 }
-static inline unsigned long get_bts_index(char *base)
+
+static inline int check_tracer(struct task_struct *task)
 {
-	return *(unsigned long *)(base + ds_cfg.bts_index.offset);
+	return (task ? (tracers >= 0) : (tracers <= 0));
 }
-static inline void set_bts_index(char *base, unsigned long value)
+
+
+/*
+ * The DS context is either attached to a thread or to a cpu:
+ * - in the former case, the thread_struct contains a pointer to the
+ *   attached context.
+ * - in the latter case, we use a static array of per-cpu context
+ *   pointers.
+ *
+ * Contexts are use-counted. They are allocated on first access and
+ * deallocated when the last user puts the context.
+ *
+ * We distinguish between an allocating and a non-allocating get of a
+ * context:
+ * - the allocating get is used for requesting BTS/PEBS resources. It
+ *   requires the caller to hold the global ds_lock.
+ * - the non-allocating get is used for all other cases. A
+ *   non-existing context indicates an error. It acquires and releases
+ *   the ds_lock itself for obtaining the context.
+ *
+ * A context and its DS configuration are allocated and deallocated
+ * together. A context always has a DS configuration of the
+ * appropriate size.
+ */
+static DEFINE_PER_CPU(struct ds_context *, system_context);
+
+#define this_system_context per_cpu(system_context, smp_processor_id())
+
+/*
+ * Returns the pointer to the parameter task's context or to the
+ * system-wide context, if task is NULL.
+ *
+ * Increases the use count of the returned context, if not NULL.
+ */
+static inline struct ds_context *ds_get_context(struct task_struct *task)
 {
-	(*(unsigned long *)(base + ds_cfg.bts_index.offset)) = value;
+	struct ds_context *context;
+
+	spin_lock(&ds_lock);
+
+	context = (task ? task->thread.ds_ctx : this_system_context);
+	if (context)
+		context->count++;
+
+	spin_unlock(&ds_lock);
+
+	return context;
 }
-static inline unsigned long get_bts_absolute_maximum(char *base)
+
+/*
+ * Same as ds_get_context, but allocates the context and it's DS
+ * structure, if necessary; returns NULL; if out of memory.
+ *
+ * pre: requires ds_lock to be held
+ */
+static inline struct ds_context *ds_alloc_context(struct task_struct *task)
 {
-	return *(unsigned long *)(base + ds_cfg.bts_absolute_maximum.offset);
+	struct ds_context **p_context =
+		(task ? &task->thread.ds_ctx : &this_system_context);
+	struct ds_context *context = *p_context;
+
+	if (!context) {
+		context = kzalloc(sizeof(*context), GFP_KERNEL);
+
+		if (!context)
+			return NULL;
+
+		context->ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL);
+		if (!context->ds) {
+			kfree(context);
+			return NULL;
+		}
+
+		*p_context = context;
+
+		context->this = p_context;
+		context->task = task;
+
+		if (task)
+			set_tsk_thread_flag(task, TIF_DS_AREA_MSR);
+
+		if (!task || (task == current))
+			wrmsr(MSR_IA32_DS_AREA, (unsigned long)context->ds, 0);
+
+		get_tracer(task);
+	}
+
+	context->count++;
+
+	return context;
 }
-static inline void set_bts_absolute_maximum(char *base, unsigned long value)
+
+/*
+ * Decreases the use count of the parameter context, if not NULL.
+ * Deallocates the context, if the use count reaches zero.
+ */
+static inline void ds_put_context(struct ds_context *context)
 {
-	(*(unsigned long *)(base + ds_cfg.bts_absolute_maximum.offset)) = value;
+	if (!context)
+		return;
+
+	spin_lock(&ds_lock);
+
+	if (--context->count)
+		goto out;
+
+	*(context->this) = NULL;
+
+	if (context->task)
+		clear_tsk_thread_flag(context->task, TIF_DS_AREA_MSR);
+
+	if (!context->task || (context->task == current))
+		wrmsrl(MSR_IA32_DS_AREA, 0);
+
+	put_tracer(context->task);
+
+	/* free any leftover buffers from tracers that did not
+	 * deallocate them properly. */
+	kfree(context->buffer[ds_bts]);
+	kfree(context->buffer[ds_pebs]);
+	kfree(context->ds);
+	kfree(context);
+ out:
+	spin_unlock(&ds_lock);
 }
-static inline unsigned long get_bts_interrupt_threshold(char *base)
+
+
+/*
+ * Handle a buffer overflow
+ *
+ * task: the task whose buffers are overflowing;
+ *       NULL for a buffer overflow on the current cpu
+ * context: the ds context
+ * qual: the buffer type
+ */
+static void ds_overflow(struct task_struct *task, struct ds_context *context,
+			enum ds_qualifier qual)
 {
-	return *(unsigned long *)(base + ds_cfg.bts_interrupt_threshold.offset);
+	if (!context)
+		return;
+
+	if (context->callback[qual])
+		(*context->callback[qual])(task);
+
+	/* todo: do some more overflow handling */
 }
-static inline void set_bts_interrupt_threshold(char *base, unsigned long value)
+
+
+/*
+ * Allocate a non-pageable buffer of the parameter size.
+ * Checks the memory and the locked memory rlimit.
+ *
+ * Returns the buffer, if successful;
+ *         NULL, if out of memory or rlimit exceeded.
+ *
+ * size: the requested buffer size in bytes
+ * pages (out): if not NULL, contains the number of pages reserved
+ */
+static inline void *ds_allocate_buffer(size_t size, unsigned int *pages)
 {
-	(*(unsigned long *)(base + ds_cfg.bts_interrupt_threshold.offset)) = value;
+	unsigned long rlim, vm, pgsz;
+	void *buffer;
+
+	pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT;
+
+	rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT;
+	vm   = current->mm->total_vm  + pgsz;
+	if (rlim < vm)
+		return NULL;
+
+	rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT;
+	vm   = current->mm->locked_vm  + pgsz;
+	if (rlim < vm)
+		return NULL;
+
+	buffer = kzalloc(size, GFP_KERNEL);
+	if (!buffer)
+		return NULL;
+
+	current->mm->total_vm  += pgsz;
+	current->mm->locked_vm += pgsz;
+
+	if (pages)
+		*pages = pgsz;
+
+	return buffer;
 }
-static inline unsigned long get_from_ip(char *base)
+
+static int ds_request(struct task_struct *task, void *base, size_t size,
+		      ds_ovfl_callback_t ovfl, enum ds_qualifier qual)
 {
-	return *(unsigned long *)(base + ds_cfg.from_ip.offset);
+	struct ds_context *context;
+	unsigned long buffer, adj;
+	const unsigned long alignment = (1 << 3);
+	int error = 0;
+
+	if (!ds_cfg.sizeof_ds)
+		return -EOPNOTSUPP;
+
+	/* we require some space to do alignment adjustments below */
+	if (size < (alignment + ds_cfg.sizeof_rec[qual]))
+		return -EINVAL;
+
+	/* buffer overflow notification is not yet implemented */
+	if (ovfl)
+		return -EOPNOTSUPP;
+
+
+	spin_lock(&ds_lock);
+
+	if (!check_tracer(task))
+		return -EPERM;
+
+	error = -ENOMEM;
+	context = ds_alloc_context(task);
+	if (!context)
+		goto out_unlock;
+
+	error = -EALREADY;
+	if (context->owner[qual] == current)
+		goto out_unlock;
+	error = -EPERM;
+	if (context->owner[qual] != NULL)
+		goto out_unlock;
+	context->owner[qual] = current;
+
+	spin_unlock(&ds_lock);
+
+
+	error = -ENOMEM;
+	if (!base) {
+		base = ds_allocate_buffer(size, &context->pages[qual]);
+		if (!base)
+			goto out_release;
+
+		context->buffer[qual]   = base;
+	}
+	error = 0;
+
+	context->callback[qual] = ovfl;
+
+	/* adjust the buffer address and size to meet alignment
+	 * constraints:
+	 * - buffer is double-word aligned
+	 * - size is multiple of record size
+	 *
+	 * We checked the size at the very beginning; we have enough
+	 * space to do the adjustment.
+	 */
+	buffer = (unsigned long)base;
+
+	adj = ALIGN(buffer, alignment) - buffer;
+	buffer += adj;
+	size   -= adj;
+
+	size /= ds_cfg.sizeof_rec[qual];
+	size *= ds_cfg.sizeof_rec[qual];
+
+	ds_set(context->ds, qual, ds_buffer_base, buffer);
+	ds_set(context->ds, qual, ds_index, buffer);
+	ds_set(context->ds, qual, ds_absolute_maximum, buffer + size);
+
+	if (ovfl) {
+		/* todo: select a suitable interrupt threshold */
+	} else
+		ds_set(context->ds, qual,
+		       ds_interrupt_threshold, buffer + size + 1);
+
+	/* we keep the context until ds_release */
+	return error;
+
+ out_release:
+	context->owner[qual] = NULL;
+	ds_put_context(context);
+	return error;
+
+ out_unlock:
+	spin_unlock(&ds_lock);
+	ds_put_context(context);
+	return error;
 }
-static inline void set_from_ip(char *base, unsigned long value)
+
+int ds_request_bts(struct task_struct *task, void *base, size_t size,
+		   ds_ovfl_callback_t ovfl)
 {
-	(*(unsigned long *)(base + ds_cfg.from_ip.offset)) = value;
+	return ds_request(task, base, size, ovfl, ds_bts);
 }
-static inline unsigned long get_to_ip(char *base)
+
+int ds_request_pebs(struct task_struct *task, void *base, size_t size,
+		    ds_ovfl_callback_t ovfl)
 {
-	return *(unsigned long *)(base + ds_cfg.to_ip.offset);
+	return ds_request(task, base, size, ovfl, ds_pebs);
 }
-static inline void set_to_ip(char *base, unsigned long value)
+
+static int ds_release(struct task_struct *task, enum ds_qualifier qual)
 {
-	(*(unsigned long *)(base + ds_cfg.to_ip.offset)) = value;
+	struct ds_context *context;
+	int error;
+
+	context = ds_get_context(task);
+	error = ds_validate_access(context, qual);
+	if (error < 0)
+		goto out;
+
+	kfree(context->buffer[qual]);
+	context->buffer[qual] = NULL;
+
+	current->mm->total_vm  -= context->pages[qual];
+	current->mm->locked_vm -= context->pages[qual];
+	context->pages[qual] = 0;
+	context->owner[qual] = NULL;
+
+	/*
+	 * we put the context twice:
+	 *   once for the ds_get_context
+	 *   once for the corresponding ds_request
+	 */
+	ds_put_context(context);
+ out:
+	ds_put_context(context);
+	return error;
 }
-static inline unsigned char get_info_type(char *base)
+
+int ds_release_bts(struct task_struct *task)
 {
-	return *(unsigned char *)(base + ds_cfg.info_type.offset);
+	return ds_release(task, ds_bts);
 }
-static inline void set_info_type(char *base, unsigned char value)
+
+int ds_release_pebs(struct task_struct *task)
 {
-	(*(unsigned char *)(base + ds_cfg.info_type.offset)) = value;
+	return ds_release(task, ds_pebs);
 }
-static inline unsigned long get_info_data(char *base)
+
+static int ds_get_index(struct task_struct *task, size_t *pos,
+			enum ds_qualifier qual)
 {
-	return *(unsigned long *)(base + ds_cfg.info_data.offset);
+	struct ds_context *context;
+	unsigned long base, index;
+	int error;
+
+	context = ds_get_context(task);
+	error = ds_validate_access(context, qual);
+	if (error < 0)
+		goto out;
+
+	base  = ds_get(context->ds, qual, ds_buffer_base);
+	index = ds_get(context->ds, qual, ds_index);
+
+	error = ((index - base) / ds_cfg.sizeof_rec[qual]);
+	if (pos)
+		*pos = error;
+ out:
+	ds_put_context(context);
+	return error;
 }
-static inline void set_info_data(char *base, unsigned long value)
+
+int ds_get_bts_index(struct task_struct *task, size_t *pos)
 {
-	(*(unsigned long *)(base + ds_cfg.info_data.offset)) = value;
+	return ds_get_index(task, pos, ds_bts);
 }
 
+int ds_get_pebs_index(struct task_struct *task, size_t *pos)
+{
+	return ds_get_index(task, pos, ds_pebs);
+}
 
-int ds_allocate(void **dsp, size_t bts_size_in_bytes)
+static int ds_get_end(struct task_struct *task, size_t *pos,
+		      enum ds_qualifier qual)
 {
-	size_t bts_size_in_records;
-	unsigned long bts;
-	void *ds;
+	struct ds_context *context;
+	unsigned long base, end;
+	int error;
+
+	context = ds_get_context(task);
+	error = ds_validate_access(context, qual);
+	if (error < 0)
+		goto out;
+
+	base = ds_get(context->ds, qual, ds_buffer_base);
+	end  = ds_get(context->ds, qual, ds_absolute_maximum);
+
+	error = ((end - base) / ds_cfg.sizeof_rec[qual]);
+	if (pos)
+		*pos = error;
+ out:
+	ds_put_context(context);
+	return error;
+}
 
-	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
-		return -EOPNOTSUPP;
+int ds_get_bts_end(struct task_struct *task, size_t *pos)
+{
+	return ds_get_end(task, pos, ds_bts);
+}
 
-	if (bts_size_in_bytes < 0)
-		return -EINVAL;
+int ds_get_pebs_end(struct task_struct *task, size_t *pos)
+{
+	return ds_get_end(task, pos, ds_pebs);
+}
 
-	bts_size_in_records =
-		bts_size_in_bytes / ds_cfg.sizeof_bts;
-	bts_size_in_bytes =
-		bts_size_in_records * ds_cfg.sizeof_bts;
+static int ds_access(struct task_struct *task, size_t index,
+		     const void **record, enum ds_qualifier qual)
+{
+	struct ds_context *context;
+	unsigned long base, idx;
+	int error;
 
-	if (bts_size_in_bytes <= 0)
+	if (!record)
 		return -EINVAL;
 
-	bts = (unsigned long)kzalloc(bts_size_in_bytes, GFP_KERNEL);
-
-	if (!bts)
-		return -ENOMEM;
+	context = ds_get_context(task);
+	error = ds_validate_access(context, qual);
+	if (error < 0)
+		goto out;
 
-	ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL);
+	base = ds_get(context->ds, qual, ds_buffer_base);
+	idx = base + (index * ds_cfg.sizeof_rec[qual]);
 
-	if (!ds) {
-		kfree((void *)bts);
-		return -ENOMEM;
-	}
-
-	set_bts_buffer_base(ds, bts);
-	set_bts_index(ds, bts);
-	set_bts_absolute_maximum(ds, bts + bts_size_in_bytes);
-	set_bts_interrupt_threshold(ds, bts + bts_size_in_bytes + 1);
+	error = -EINVAL;
+	if (idx > ds_get(context->ds, qual, ds_absolute_maximum))
+		goto out;
 
-	*dsp = ds;
-	return 0;
+	*record = (const void *)idx;
+	error = ds_cfg.sizeof_rec[qual];
+ out:
+	ds_put_context(context);
+	return error;
 }
 
-int ds_free(void **dsp)
+int ds_access_bts(struct task_struct *task, size_t index, const void **record)
 {
-	if (*dsp) {
-		kfree((void *)get_bts_buffer_base(*dsp));
-		kfree(*dsp);
-		*dsp = NULL;
-	}
-	return 0;
+	return ds_access(task, index, record, ds_bts);
 }
 
-int ds_get_bts_size(void *ds)
+int ds_access_pebs(struct task_struct *task, size_t index, const void **record)
 {
-	int size_in_bytes;
-
-	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
-		return -EOPNOTSUPP;
-
-	if (!ds)
-		return 0;
-
-	size_in_bytes =
-		get_bts_absolute_maximum(ds) -
-		get_bts_buffer_base(ds);
-	return size_in_bytes;
+	return ds_access(task, index, record, ds_pebs);
 }
 
-int ds_get_bts_end(void *ds)
+static int ds_write(struct task_struct *task, const void *record, size_t size,
+		    enum ds_qualifier qual, int force)
 {
-	int size_in_bytes = ds_get_bts_size(ds);
-
-	if (size_in_bytes <= 0)
-		return size_in_bytes;
+	struct ds_context *context;
+	int error;
 
-	return size_in_bytes / ds_cfg.sizeof_bts;
-}
+	if (!record)
+		return -EINVAL;
 
-int ds_get_bts_index(void *ds)
-{
-	int index_offset_in_bytes;
+	error = -EPERM;
+	context = ds_get_context(task);
+	if (!context)
+		goto out;
 
-	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
-		return -EOPNOTSUPP;
+	if (!force) {
+		error = ds_validate_access(context, qual);
+		if (error < 0)
+			goto out;
+	}
 
-	index_offset_in_bytes =
-		get_bts_index(ds) -
-		get_bts_buffer_base(ds);
+	error = 0;
+	while (size) {
+		unsigned long base, index, end, write_end, int_th;
+		unsigned long write_size, adj_write_size;
+
+		/*
+		 * write as much as possible without producing an
+		 * overflow interrupt.
+		 *
+		 * interrupt_threshold must either be
+		 * - bigger than absolute_maximum or
+		 * - point to a record between buffer_base and absolute_maximum
+		 *
+		 * index points to a valid record.
+		 */
+		base   = ds_get(context->ds, qual, ds_buffer_base);
+		index  = ds_get(context->ds, qual, ds_index);
+		end    = ds_get(context->ds, qual, ds_absolute_maximum);
+		int_th = ds_get(context->ds, qual, ds_interrupt_threshold);
+
+		write_end = min(end, int_th);
+
+		/* if we are already beyond the interrupt threshold,
+		 * we fill the entire buffer */
+		if (write_end <= index)
+			write_end = end;
+
+		if (write_end <= index)
+			goto out;
+
+		write_size = min((unsigned long) size, write_end - index);
+		memcpy((void *)index, record, write_size);
+
+		record = (const char *)record + write_size;
+		size  -= write_size;
+		error += write_size;
+
+		adj_write_size = write_size / ds_cfg.sizeof_rec[qual];
+		adj_write_size *= ds_cfg.sizeof_rec[qual];
+
+		/* zero out trailing bytes */
+		memset((char *)index + write_size, 0,
+		       adj_write_size - write_size);
+		index += adj_write_size;
+
+		if (index >= end)
+			index = base;
+		ds_set(context->ds, qual, ds_index, index);
+
+		if (index >= int_th)
+			ds_overflow(task, context, qual);
+	}
 
-	return index_offset_in_bytes / ds_cfg.sizeof_bts;
+ out:
+	ds_put_context(context);
+	return error;
 }
 
-int ds_set_overflow(void *ds, int method)
+int ds_write_bts(struct task_struct *task, const void *record, size_t size)
 {
-	switch (method) {
-	case DS_O_SIGNAL:
-		return -EOPNOTSUPP;
-	case DS_O_WRAP:
-		return 0;
-	default:
-		return -EINVAL;
-	}
+	return ds_write(task, record, size, ds_bts, /* force = */ 0);
 }
 
-int ds_get_overflow(void *ds)
+int ds_write_pebs(struct task_struct *task, const void *record, size_t size)
 {
-	return DS_O_WRAP;
+	return ds_write(task, record, size, ds_pebs, /* force = */ 0);
 }
 
-int ds_clear(void *ds)
+int ds_unchecked_write_bts(struct task_struct *task,
+			   const void *record, size_t size)
 {
-	int bts_size = ds_get_bts_size(ds);
-	unsigned long bts_base;
-
-	if (bts_size <= 0)
-		return bts_size;
-
-	bts_base = get_bts_buffer_base(ds);
-	memset((void *)bts_base, 0, bts_size);
-
-	set_bts_index(ds, bts_base);
-	return 0;
+	return ds_write(task, record, size, ds_bts, /* force = */ 1);
 }
 
-int ds_read_bts(void *ds, int index, struct bts_struct *out)
+int ds_unchecked_write_pebs(struct task_struct *task,
+			    const void *record, size_t size)
 {
-	void *bts;
+	return ds_write(task, record, size, ds_pebs, /* force = */ 1);
+}
 
-	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
-		return -EOPNOTSUPP;
+static int ds_reset_or_clear(struct task_struct *task,
+			     enum ds_qualifier qual, int clear)
+{
+	struct ds_context *context;
+	unsigned long base, end;
+	int error;
 
-	if (index < 0)
-		return -EINVAL;
+	context = ds_get_context(task);
+	error = ds_validate_access(context, qual);
+	if (error < 0)
+		goto out;
 
-	if (index >= ds_get_bts_size(ds))
-		return -EINVAL;
+	base = ds_get(context->ds, qual, ds_buffer_base);
+	end  = ds_get(context->ds, qual, ds_absolute_maximum);
 
-	bts = (void *)(get_bts_buffer_base(ds) + (index * ds_cfg.sizeof_bts));
+	if (clear)
+		memset((void *)base, 0, end - base);
 
-	memset(out, 0, sizeof(*out));
-	if (get_from_ip(bts) == BTS_ESCAPE_ADDRESS) {
-		out->qualifier       = get_info_type(bts);
-		out->variant.jiffies = get_info_data(bts);
-	} else {
-		out->qualifier = BTS_BRANCH;
-		out->variant.lbr.from_ip = get_from_ip(bts);
-		out->variant.lbr.to_ip   = get_to_ip(bts);
-	}
+	ds_set(context->ds, qual, ds_index, base);
 
-	return sizeof(*out);;
+	error = 0;
+ out:
+	ds_put_context(context);
+	return error;
 }
 
-int ds_write_bts(void *ds, const struct bts_struct *in)
+int ds_reset_bts(struct task_struct *task)
 {
-	unsigned long bts;
-
-	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
-		return -EOPNOTSUPP;
-
-	if (ds_get_bts_size(ds) <= 0)
-		return -ENXIO;
+	return ds_reset_or_clear(task, ds_bts, /* clear = */ 0);
+}
 
-	bts = get_bts_index(ds);
+int ds_reset_pebs(struct task_struct *task)
+{
+	return ds_reset_or_clear(task, ds_pebs, /* clear = */ 0);
+}
 
-	memset((void *)bts, 0, ds_cfg.sizeof_bts);
-	switch (in->qualifier) {
-	case BTS_INVALID:
-		break;
+int ds_clear_bts(struct task_struct *task)
+{
+	return ds_reset_or_clear(task, ds_bts, /* clear = */ 1);
+}
 
-	case BTS_BRANCH:
-		set_from_ip((void *)bts, in->variant.lbr.from_ip);
-		set_to_ip((void *)bts, in->variant.lbr.to_ip);
-		break;
+int ds_clear_pebs(struct task_struct *task)
+{
+	return ds_reset_or_clear(task, ds_pebs, /* clear = */ 1);
+}
 
-	case BTS_TASK_ARRIVES:
-	case BTS_TASK_DEPARTS:
-		set_from_ip((void *)bts, BTS_ESCAPE_ADDRESS);
-		set_info_type((void *)bts, in->qualifier);
-		set_info_data((void *)bts, in->variant.jiffies);
-		break;
+int ds_get_pebs_reset(struct task_struct *task, u64 *value)
+{
+	struct ds_context *context;
+	int error;
 
-	default:
+	if (!value)
 		return -EINVAL;
-	}
 
-	bts = bts + ds_cfg.sizeof_bts;
-	if (bts >= get_bts_absolute_maximum(ds))
-		bts = get_bts_buffer_base(ds);
-	set_bts_index(ds, bts);
+	context = ds_get_context(task);
+	error = ds_validate_access(context, ds_pebs);
+	if (error < 0)
+		goto out;
 
-	return ds_cfg.sizeof_bts;
+	*value = *(u64 *)(context->ds + (ds_cfg.sizeof_field * 8));
+
+	error = 0;
+ out:
+	ds_put_context(context);
+	return error;
 }
 
-unsigned long ds_debugctl_mask(void)
+int ds_set_pebs_reset(struct task_struct *task, u64 value)
 {
-	return ds_cfg.debugctl_mask;
-}
+	struct ds_context *context;
+	int error;
 
-#ifdef __i386__
-static const struct ds_configuration ds_cfg_netburst = {
-	.sizeof_ds = 9 * 4,
-	.bts_buffer_base = { 0, 4 },
-	.bts_index = { 4, 4 },
-	.bts_absolute_maximum = { 8, 4 },
-	.bts_interrupt_threshold = { 12, 4 },
-	.sizeof_bts = 3 * 4,
-	.from_ip = { 0, 4 },
-	.to_ip = { 4, 4 },
-	.info_type = { 4, 1 },
-	.info_data = { 8, 4 },
-	.debugctl_mask = (1<<2)|(1<<3)
-};
+	context = ds_get_context(task);
+	error = ds_validate_access(context, ds_pebs);
+	if (error < 0)
+		goto out;
 
-static const struct ds_configuration ds_cfg_pentium_m = {
-	.sizeof_ds = 9 * 4,
-	.bts_buffer_base = { 0, 4 },
-	.bts_index = { 4, 4 },
-	.bts_absolute_maximum = { 8, 4 },
-	.bts_interrupt_threshold = { 12, 4 },
-	.sizeof_bts = 3 * 4,
-	.from_ip = { 0, 4 },
-	.to_ip = { 4, 4 },
-	.info_type = { 4, 1 },
-	.info_data = { 8, 4 },
-	.debugctl_mask = (1<<6)|(1<<7)
+	*(u64 *)(context->ds + (ds_cfg.sizeof_field * 8)) = value;
+
+	error = 0;
+ out:
+	ds_put_context(context);
+	return error;
+}
+
+static const struct ds_configuration ds_cfg_var = {
+	.sizeof_ds    = sizeof(long) * 12,
+	.sizeof_field = sizeof(long),
+	.sizeof_rec[ds_bts]   = sizeof(long) * 3,
+	.sizeof_rec[ds_pebs]  = sizeof(long) * 10
 };
-#endif /* _i386_ */
-
-static const struct ds_configuration ds_cfg_core2 = {
-	.sizeof_ds = 9 * 8,
-	.bts_buffer_base = { 0, 8 },
-	.bts_index = { 8, 8 },
-	.bts_absolute_maximum = { 16, 8 },
-	.bts_interrupt_threshold = { 24, 8 },
-	.sizeof_bts = 3 * 8,
-	.from_ip = { 0, 8 },
-	.to_ip = { 8, 8 },
-	.info_type = { 8, 1 },
-	.info_data = { 16, 8 },
-	.debugctl_mask = (1<<6)|(1<<7)|(1<<9)
+static const struct ds_configuration ds_cfg_64 = {
+	.sizeof_ds    = 8 * 12,
+	.sizeof_field = 8,
+	.sizeof_rec[ds_bts]   = 8 * 3,
+	.sizeof_rec[ds_pebs]  = 8 * 10
 };
 
 static inline void
@@ -429,14 +821,13 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
 	switch (c->x86) {
 	case 0x6:
 		switch (c->x86_model) {
-#ifdef __i386__
 		case 0xD:
 		case 0xE: /* Pentium M */
-			ds_configure(&ds_cfg_pentium_m);
+			ds_configure(&ds_cfg_var);
 			break;
-#endif /* _i386_ */
 		case 0xF: /* Core2 */
-			ds_configure(&ds_cfg_core2);
+		case 0x1C: /* Atom */
+			ds_configure(&ds_cfg_64);
 			break;
 		default:
 			/* sorry, don't know about them */
@@ -445,13 +836,11 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
 		break;
 	case 0xF:
 		switch (c->x86_model) {
-#ifdef __i386__
 		case 0x0:
 		case 0x1:
 		case 0x2: /* Netburst */
-			ds_configure(&ds_cfg_netburst);
+			ds_configure(&ds_cfg_var);
 			break;
-#endif /* _i386_ */
 		default:
 			/* sorry, don't know about them */
 			break;
@@ -462,3 +851,14 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
 		break;
 	}
 }
+
+void ds_free(struct ds_context *context)
+{
+	/* This is called when the task owning the parameter context
+	 * is dying. There should not be any user of that context left
+	 * to disturb us, anymore. */
+	unsigned long leftovers = context->count;
+	while (leftovers--)
+		ds_put_context(context);
+}
+#endif /* CONFIG_X86_DS */
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index e24d1bc47b4..78e642feac3 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -1206,7 +1206,7 @@ static int __init parse_memmap_opt(char *p)
 	if (!p)
 		return -EINVAL;
 
-	if (!strcmp(p, "exactmap")) {
+	if (!strncmp(p, "exactmap", 8)) {
 #ifdef CONFIG_CRASH_DUMP
 		/*
 		 * If we are doing a crash dump, we still need to know
diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index 4353cf5e6fa..24bb5faf5ef 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -95,6 +95,20 @@ static void __init nvidia_bugs(int num, int slot, int func)
 
 }
 
+#ifdef CONFIG_DMAR
+static void __init intel_g33_dmar(int num, int slot, int func)
+{
+	struct acpi_table_header *dmar_tbl;
+	acpi_status status;
+
+	status = acpi_get_table(ACPI_SIG_DMAR, 0, &dmar_tbl);
+	if (ACPI_SUCCESS(status)) {
+		printk(KERN_INFO "BIOS BUG: DMAR advertised on Intel G31/G33 chipset -- ignoring\n");
+		dmar_disabled = 1;
+	}
+}
+#endif
+
 #define QFLAG_APPLY_ONCE 	0x1
 #define QFLAG_APPLIED		0x2
 #define QFLAG_DONE		(QFLAG_APPLY_ONCE|QFLAG_APPLIED)
@@ -114,6 +128,10 @@ static struct chipset early_qrk[] __initdata = {
 	  PCI_CLASS_BRIDGE_PCI, PCI_ANY_ID, QFLAG_APPLY_ONCE, via_bugs },
 	{ PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB,
 	  PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, fix_hypertransport_config },
+#ifdef CONFIG_DMAR
+	{ PCI_VENDOR_ID_INTEL, 0x29c0,
+	  PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, intel_g33_dmar },
+#endif
 	{}
 };
 
diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index 06cc8d4254b..945a31cdd81 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -414,9 +414,11 @@ void __init efi_init(void)
 	if (memmap.map == NULL)
 		printk(KERN_ERR "Could not map the EFI memory map!\n");
 	memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size);
+
 	if (memmap.desc_size != sizeof(efi_memory_desc_t))
-		printk(KERN_WARNING "Kernel-defined memdesc"
-		       "doesn't match the one from EFI!\n");
+		printk(KERN_WARNING
+		  "Kernel-defined memdesc doesn't match the one from EFI!\n");
+
 	if (add_efi_memmap)
 		do_add_efi_memmap();
 
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 9bfc4d72fb2..d16084f9064 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -108,12 +108,11 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	}
 	load_idt((const struct desc_ptr *)&idt_descr);
 
-	early_printk("Kernel alive\n");
+	if (console_loglevel == 10)
+		early_printk("Kernel alive\n");
 
 	x86_64_init_pda();
 
-	early_printk("Kernel really alive\n");
-
 	x86_64_start_reservations(real_mode_data);
 }
 
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index a7010c3a377..e835b4eea70 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -172,10 +172,6 @@ num_subarch_entries = (. - subarch_entries) / 4
  *
  * Note that the stack is not yet set up!
  */
-#define PTE_ATTR	0x007		/* PRESENT+RW+USER */
-#define PDE_ATTR	0x067		/* PRESENT+RW+USER+DIRTY+ACCESSED */
-#define PGD_ATTR	0x001		/* PRESENT (no other attributes) */
-
 default_entry:
 #ifdef CONFIG_X86_PAE
 
@@ -196,9 +192,9 @@ default_entry:
 	movl $pa(pg0), %edi
 	movl %edi, pa(init_pg_tables_start)
 	movl $pa(swapper_pg_pmd), %edx
-	movl $PTE_ATTR, %eax
+	movl $PTE_IDENT_ATTR, %eax
 10:
-	leal PDE_ATTR(%edi),%ecx		/* Create PMD entry */
+	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PMD entry */
 	movl %ecx,(%edx)			/* Store PMD entry */
 						/* Upper half already zero */
 	addl $8,%edx
@@ -215,7 +211,7 @@ default_entry:
 	 * End condition: we must map up to and including INIT_MAP_BEYOND_END
 	 * bytes beyond the end of our own page tables.
 	 */
-	leal (INIT_MAP_BEYOND_END+PTE_ATTR)(%edi),%ebp
+	leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp
 	cmpl %ebp,%eax
 	jb 10b
 1:
@@ -224,7 +220,7 @@ default_entry:
 	movl %eax, pa(max_pfn_mapped)
 
 	/* Do early initialization of the fixmap area */
-	movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax
+	movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax
 	movl %eax,pa(swapper_pg_pmd+0x1000*KPMDS-8)
 #else	/* Not PAE */
 
@@ -233,9 +229,9 @@ page_pde_offset = (__PAGE_OFFSET >> 20);
 	movl $pa(pg0), %edi
 	movl %edi, pa(init_pg_tables_start)
 	movl $pa(swapper_pg_dir), %edx
-	movl $PTE_ATTR, %eax
+	movl $PTE_IDENT_ATTR, %eax
 10:
-	leal PDE_ATTR(%edi),%ecx		/* Create PDE entry */
+	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PDE entry */
 	movl %ecx,(%edx)			/* Store identity PDE entry */
 	movl %ecx,page_pde_offset(%edx)		/* Store kernel PDE entry */
 	addl $4,%edx
@@ -249,7 +245,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20);
 	 * bytes beyond the end of our own page tables; the +0x007 is
 	 * the attribute bits
 	 */
-	leal (INIT_MAP_BEYOND_END+PTE_ATTR)(%edi),%ebp
+	leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp
 	cmpl %ebp,%eax
 	jb 10b
 	movl %edi,pa(init_pg_tables_end)
@@ -257,7 +253,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20);
 	movl %eax, pa(max_pfn_mapped)
 
 	/* Do early initialization of the fixmap area */
-	movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax
+	movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax
 	movl %eax,pa(swapper_pg_dir+0xffc)
 #endif
 	jmp 3f
@@ -634,19 +630,19 @@ ENTRY(empty_zero_page)
 	/* Page-aligned for the benefit of paravirt? */
 	.align PAGE_SIZE_asm
 ENTRY(swapper_pg_dir)
-	.long	pa(swapper_pg_pmd+PGD_ATTR),0		/* low identity map */
+	.long	pa(swapper_pg_pmd+PGD_IDENT_ATTR),0	/* low identity map */
 # if KPMDS == 3
-	.long	pa(swapper_pg_pmd+PGD_ATTR),0
-	.long	pa(swapper_pg_pmd+PGD_ATTR+0x1000),0
-	.long	pa(swapper_pg_pmd+PGD_ATTR+0x2000),0
+	.long	pa(swapper_pg_pmd+PGD_IDENT_ATTR),0
+	.long	pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0
+	.long	pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x2000),0
 # elif KPMDS == 2
 	.long	0,0
-	.long	pa(swapper_pg_pmd+PGD_ATTR),0
-	.long	pa(swapper_pg_pmd+PGD_ATTR+0x1000),0
+	.long	pa(swapper_pg_pmd+PGD_IDENT_ATTR),0
+	.long	pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0
 # elif KPMDS == 1
 	.long	0,0
 	.long	0,0
-	.long	pa(swapper_pg_pmd+PGD_ATTR),0
+	.long	pa(swapper_pg_pmd+PGD_IDENT_ATTR),0
 # else
 #  error "Kernel PMDs should be 1, 2 or 3"
 # endif
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index db3280afe88..26cfdc1d7c7 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -110,7 +110,7 @@ startup_64:
 	movq	%rdi, %rax
 	shrq	$PMD_SHIFT, %rax
 	andq	$(PTRS_PER_PMD - 1), %rax
-	leaq	__PAGE_KERNEL_LARGE_EXEC(%rdi), %rdx
+	leaq	__PAGE_KERNEL_IDENT_LARGE_EXEC(%rdi), %rdx
 	leaq	level2_spare_pgt(%rip), %rbx
 	movq	%rdx, 0(%rbx, %rax, 8)
 ident_complete:
@@ -374,7 +374,7 @@ NEXT_PAGE(level2_ident_pgt)
 	/* Since I easily can, map the first 1G.
 	 * Don't set NX because code runs from these pages.
 	 */
-	PMDS(0, __PAGE_KERNEL_LARGE_EXEC, PTRS_PER_PMD)
+	PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD)
 
 NEXT_PAGE(level2_kernel_pgt)
 	/*
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 59fd3b6b130..73deaffadd0 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -210,8 +210,8 @@ static void hpet_legacy_clockevent_register(void)
 	/* Calculate the min / max delta */
 	hpet_clockevent.max_delta_ns = clockevent_delta2ns(0x7FFFFFFF,
 							   &hpet_clockevent);
-	hpet_clockevent.min_delta_ns = clockevent_delta2ns(0x30,
-							   &hpet_clockevent);
+	/* 5 usec minimum reprogramming delta. */
+	hpet_clockevent.min_delta_ns = 5000;
 
 	/*
 	 * Start hpet with the boot cpu mask and make it
@@ -270,15 +270,22 @@ static void hpet_legacy_set_mode(enum clock_event_mode mode,
 }
 
 static int hpet_legacy_next_event(unsigned long delta,
-			   struct clock_event_device *evt)
+				  struct clock_event_device *evt)
 {
-	unsigned long cnt;
+	u32 cnt;
 
 	cnt = hpet_readl(HPET_COUNTER);
-	cnt += delta;
+	cnt += (u32) delta;
 	hpet_writel(cnt, HPET_T0_CMP);
 
-	return ((long)(hpet_readl(HPET_COUNTER) - cnt ) > 0) ? -ETIME : 0;
+	/*
+	 * We need to read back the CMP register to make sure that
+	 * what we wrote hit the chip before we compare it to the
+	 * counter.
+	 */
+	WARN_ON((u32)hpet_readl(HPET_T0_CMP) != cnt);
+
+	return (s32)((u32)hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0;
 }
 
 /*
diff --git a/arch/x86/kernel/io_delay.c b/arch/x86/kernel/io_delay.c
index 1c3a66a67f8..720d2607aac 100644
--- a/arch/x86/kernel/io_delay.c
+++ b/arch/x86/kernel/io_delay.c
@@ -92,6 +92,14 @@ static struct dmi_system_id __initdata io_delay_0xed_port_dmi_table[] = {
 			DMI_MATCH(DMI_BOARD_NAME, "30BF")
 		}
 	},
+	{
+		.callback	= dmi_io_delay_0xed_port,
+		.ident		= "Presario F700",
+		.matches	= {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
+			DMI_MATCH(DMI_BOARD_NAME, "30D3")
+		}
+	},
 	{ }
 };
 
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 1cf8c1fcc08..b71e02d42f4 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -325,7 +325,7 @@ skip:
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ",
 				per_cpu(irq_stat,j).irq_call_count);
-		seq_printf(p, "  function call interrupts\n");
+		seq_printf(p, "  Function call interrupts\n");
 		seq_printf(p, "TLB: ");
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ",
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index 1f78b238d8d..f065fe9071b 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -129,7 +129,7 @@ skip:
 		seq_printf(p, "CAL: ");
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", cpu_pda(j)->irq_call_count);
-		seq_printf(p, "  function call interrupts\n");
+		seq_printf(p, "  Function call interrupts\n");
 		seq_printf(p, "TLB: ");
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", cpu_pda(j)->irq_tlb_count);
diff --git a/arch/x86/kernel/k8.c b/arch/x86/kernel/k8.c
index 7377ccb2133..304d8bad655 100644
--- a/arch/x86/kernel/k8.c
+++ b/arch/x86/kernel/k8.c
@@ -16,8 +16,9 @@ EXPORT_SYMBOL(num_k8_northbridges);
 static u32 *flush_words;
 
 struct pci_device_id k8_nb_ids[] = {
-	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1103) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1203) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_11H_NB_MISC) },
 	{}
 };
 EXPORT_SYMBOL(k8_nb_ids);
diff --git a/arch/x86/kernel/kdebugfs.c b/arch/x86/kernel/kdebugfs.c
index f2d43bc7551..ff7d3b0124f 100644
--- a/arch/x86/kernel/kdebugfs.c
+++ b/arch/x86/kernel/kdebugfs.c
@@ -139,6 +139,7 @@ static int __init create_setup_data_nodes(struct dentry *parent)
 		if (PageHighMem(pg)) {
 			data = ioremap_cache(pa_data, sizeof(*data));
 			if (!data) {
+				kfree(node);
 				error = -ENXIO;
 				goto err_dir;
 			}
diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index f47f0eb886b..10435a120d2 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -69,6 +69,9 @@ static int gdb_x86vector = -1;
  */
 void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 {
+#ifndef CONFIG_X86_32
+	u32 *gdb_regs32 = (u32 *)gdb_regs;
+#endif
 	gdb_regs[GDB_AX]	= regs->ax;
 	gdb_regs[GDB_BX]	= regs->bx;
 	gdb_regs[GDB_CX]	= regs->cx;
@@ -76,9 +79,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 	gdb_regs[GDB_SI]	= regs->si;
 	gdb_regs[GDB_DI]	= regs->di;
 	gdb_regs[GDB_BP]	= regs->bp;
-	gdb_regs[GDB_PS]	= regs->flags;
 	gdb_regs[GDB_PC]	= regs->ip;
 #ifdef CONFIG_X86_32
+	gdb_regs[GDB_PS]	= regs->flags;
 	gdb_regs[GDB_DS]	= regs->ds;
 	gdb_regs[GDB_ES]	= regs->es;
 	gdb_regs[GDB_CS]	= regs->cs;
@@ -94,6 +97,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 	gdb_regs[GDB_R13]	= regs->r13;
 	gdb_regs[GDB_R14]	= regs->r14;
 	gdb_regs[GDB_R15]	= regs->r15;
+	gdb_regs32[GDB_PS]	= regs->flags;
+	gdb_regs32[GDB_CS]	= regs->cs;
+	gdb_regs32[GDB_SS]	= regs->ss;
 #endif
 	gdb_regs[GDB_SP]	= regs->sp;
 }
@@ -112,6 +118,9 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
  */
 void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
 {
+#ifndef CONFIG_X86_32
+	u32 *gdb_regs32 = (u32 *)gdb_regs;
+#endif
 	gdb_regs[GDB_AX]	= 0;
 	gdb_regs[GDB_BX]	= 0;
 	gdb_regs[GDB_CX]	= 0;
@@ -129,8 +138,10 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
 	gdb_regs[GDB_FS]	= 0xFFFF;
 	gdb_regs[GDB_GS]	= 0xFFFF;
 #else
-	gdb_regs[GDB_PS]	= *(unsigned long *)(p->thread.sp + 8);
-	gdb_regs[GDB_PC]	= 0;
+	gdb_regs32[GDB_PS]	= *(unsigned long *)(p->thread.sp + 8);
+	gdb_regs32[GDB_CS]	= __KERNEL_CS;
+	gdb_regs32[GDB_SS]	= __KERNEL_DS;
+	gdb_regs[GDB_PC]	= p->thread.ip;
 	gdb_regs[GDB_R8]	= 0;
 	gdb_regs[GDB_R9]	= 0;
 	gdb_regs[GDB_R10]	= 0;
@@ -153,6 +164,9 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
  */
 void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 {
+#ifndef CONFIG_X86_32
+	u32 *gdb_regs32 = (u32 *)gdb_regs;
+#endif
 	regs->ax		= gdb_regs[GDB_AX];
 	regs->bx		= gdb_regs[GDB_BX];
 	regs->cx		= gdb_regs[GDB_CX];
@@ -160,9 +174,9 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 	regs->si		= gdb_regs[GDB_SI];
 	regs->di		= gdb_regs[GDB_DI];
 	regs->bp		= gdb_regs[GDB_BP];
-	regs->flags		= gdb_regs[GDB_PS];
 	regs->ip		= gdb_regs[GDB_PC];
 #ifdef CONFIG_X86_32
+	regs->flags		= gdb_regs[GDB_PS];
 	regs->ds		= gdb_regs[GDB_DS];
 	regs->es		= gdb_regs[GDB_ES];
 	regs->cs		= gdb_regs[GDB_CS];
@@ -175,6 +189,9 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 	regs->r13		= gdb_regs[GDB_R13];
 	regs->r14		= gdb_regs[GDB_R14];
 	regs->r15		= gdb_regs[GDB_R15];
+	regs->flags		= gdb_regs32[GDB_PS];
+	regs->cs		= gdb_regs32[GDB_CS];
+	regs->ss		= gdb_regs32[GDB_SS];
 #endif
 }
 
@@ -378,10 +395,8 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code,
 		if (remcomInBuffer[0] == 's') {
 			linux_regs->flags |= X86_EFLAGS_TF;
 			kgdb_single_step = 1;
-			if (kgdb_contthread) {
-				atomic_set(&kgdb_cpu_doing_single_step,
-					   raw_smp_processor_id());
-			}
+			atomic_set(&kgdb_cpu_doing_single_step,
+				   raw_smp_processor_id());
 		}
 
 		get_debugreg(dr6, 6);
@@ -440,12 +455,7 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd)
 		return NOTIFY_DONE;
 
 	case DIE_NMI_IPI:
-		if (atomic_read(&kgdb_active) != -1) {
-			/* KGDB CPU roundup */
-			kgdb_nmicallback(raw_smp_processor_id(), regs);
-			was_in_debug_nmi[raw_smp_processor_id()] = 1;
-			touch_nmi_watchdog();
-		}
+		/* Just ignore, we will handle the roundup on DIE_NMI. */
 		return NOTIFY_DONE;
 
 	case DIE_NMIUNKNOWN:
@@ -466,9 +476,15 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd)
 
 	case DIE_DEBUG:
 		if (atomic_read(&kgdb_cpu_doing_single_step) ==
-			raw_smp_processor_id() &&
-			user_mode(regs))
-			return single_step_cont(regs, args);
+		    raw_smp_processor_id()) {
+			if (user_mode(regs))
+				return single_step_cont(regs, args);
+			break;
+		} else if (test_thread_flag(TIF_SINGLESTEP))
+			/* This means a user thread is single stepping
+			 * a system call which should be ignored
+			 */
+			return NOTIFY_DONE;
 		/* fall through */
 	default:
 		if (user_mode(regs))
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 8b7a3cf37d2..478bca986ec 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -178,7 +178,7 @@ static void kvm_flush_tlb(void)
 	kvm_deferred_mmu_op(&ftlb, sizeof ftlb);
 }
 
-static void kvm_release_pt(u32 pfn)
+static void kvm_release_pt(unsigned long pfn)
 {
 	struct kvm_mmu_op_release_pt rpt = {
 		.header.op = KVM_MMU_OP_RELEASE_PT,
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index abb78a2cc4a..2c97f07f1c2 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -299,6 +299,15 @@ void acpi_nmi_disable(void)
 		on_each_cpu(__acpi_nmi_disable, NULL, 1);
 }
 
+/*
+ * This function is called as soon the LAPIC NMI watchdog driver has everything
+ * in place and it's ready to check if the NMIs belong to the NMI watchdog
+ */
+void cpu_nmi_set_wd_enabled(void)
+{
+	__get_cpu_var(wd_enabled) = 1;
+}
+
 void setup_apic_nmi_watchdog(void *unused)
 {
 	if (__get_cpu_var(wd_enabled))
@@ -311,8 +320,6 @@ void setup_apic_nmi_watchdog(void *unused)
 
 	switch (nmi_watchdog) {
 	case NMI_LOCAL_APIC:
-		 /* enable it before to avoid race with handler */
-		__get_cpu_var(wd_enabled) = 1;
 		if (lapic_watchdog_init(nmi_hz) < 0) {
 			__get_cpu_var(wd_enabled) = 0;
 			return;
diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c
index 3e667227480..7a13fac63a1 100644
--- a/arch/x86/kernel/olpc.c
+++ b/arch/x86/kernel/olpc.c
@@ -190,12 +190,12 @@ EXPORT_SYMBOL_GPL(olpc_ec_cmd);
 static void __init platform_detect(void)
 {
 	size_t propsize;
-	u32 rev;
+	__be32 rev;
 
 	if (ofw("getprop", 4, 1, NULL, "board-revision-int", &rev, 4,
 			&propsize) || propsize != 4) {
 		printk(KERN_ERR "ofw: getprop call failed!\n");
-		rev = 0;
+		rev = cpu_to_be32(0);
 	}
 	olpc_platform_info.boardrev = be32_to_cpu(rev);
 }
@@ -203,7 +203,7 @@ static void __init platform_detect(void)
 static void __init platform_detect(void)
 {
 	/* stopgap until OFW support is added to the kernel */
-	olpc_platform_info.boardrev = be32_to_cpu(0xc2);
+	olpc_platform_info.boardrev = 0xc2;
 }
 #endif
 
diff --git a/arch/x86/kernel/paravirt_patch_32.c b/arch/x86/kernel/paravirt_patch_32.c
index 58262218781..9fe644f4861 100644
--- a/arch/x86/kernel/paravirt_patch_32.c
+++ b/arch/x86/kernel/paravirt_patch_32.c
@@ -23,7 +23,7 @@ unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
 			start = start_##ops##_##x;		\
 			end = end_##ops##_##x;			\
 			goto patch_site
-	switch(type) {
+	switch (type) {
 		PATCH_SITE(pv_irq_ops, irq_disable);
 		PATCH_SITE(pv_irq_ops, irq_enable);
 		PATCH_SITE(pv_irq_ops, restore_fl);
diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index dcdac6c826e..080d1d27f37 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -261,7 +261,7 @@ static void iommu_range_reserve(struct iommu_table *tbl,
 			       badbit, tbl, start_addr, npages);
 	}
 
-	set_bit_string(tbl->it_map, index, npages);
+	iommu_area_reserve(tbl->it_map, index, npages);
 
 	spin_unlock_irqrestore(&tbl->it_lock, flags);
 }
@@ -491,6 +491,8 @@ static void* calgary_alloc_coherent(struct device *dev, size_t size,
 	npages = size >> PAGE_SHIFT;
 	order = get_order(size);
 
+	flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
+
 	/* alloc enough pages (and possibly more) */
 	ret = (void *)__get_free_pages(flag, order);
 	if (!ret)
@@ -510,8 +512,22 @@ error:
 	return ret;
 }
 
+static void calgary_free_coherent(struct device *dev, size_t size,
+				  void *vaddr, dma_addr_t dma_handle)
+{
+	unsigned int npages;
+	struct iommu_table *tbl = find_iommu_table(dev);
+
+	size = PAGE_ALIGN(size);
+	npages = size >> PAGE_SHIFT;
+
+	iommu_free(tbl, dma_handle, npages);
+	free_pages((unsigned long)vaddr, get_order(size));
+}
+
 static struct dma_mapping_ops calgary_dma_ops = {
 	.alloc_coherent = calgary_alloc_coherent,
+	.free_coherent = calgary_free_coherent,
 	.map_single = calgary_map_single,
 	.unmap_single = calgary_unmap_single,
 	.map_sg = calgary_map_sg,
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 87d4d6964ec..0a3824e837b 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -41,11 +41,12 @@ EXPORT_SYMBOL(bad_dma_address);
 /* Dummy device used for NULL arguments (normally ISA). Better would
    be probably a smaller DMA mask, but this is bug-to-bug compatible
    to older i386. */
-struct device fallback_dev = {
+struct device x86_dma_fallback_dev = {
 	.bus_id = "fallback device",
 	.coherent_dma_mask = DMA_32BIT_MASK,
-	.dma_mask = &fallback_dev.coherent_dma_mask,
+	.dma_mask = &x86_dma_fallback_dev.coherent_dma_mask,
 };
+EXPORT_SYMBOL(x86_dma_fallback_dev);
 
 int dma_set_mask(struct device *dev, u64 mask)
 {
@@ -82,7 +83,7 @@ void __init dma32_reserve_bootmem(void)
 	 * using 512M as goal
 	 */
 	align = 64ULL<<20;
-	size = round_up(dma32_bootmem_size, align);
+	size = roundup(dma32_bootmem_size, align);
 	dma32_bootmem_ptr = __alloc_bootmem_nopanic(size, align,
 				 512ULL<<20);
 	if (dma32_bootmem_ptr)
@@ -133,6 +134,37 @@ unsigned long iommu_num_pages(unsigned long addr, unsigned long len)
 EXPORT_SYMBOL(iommu_num_pages);
 #endif
 
+void *dma_generic_alloc_coherent(struct device *dev, size_t size,
+				 dma_addr_t *dma_addr, gfp_t flag)
+{
+	unsigned long dma_mask;
+	struct page *page;
+	dma_addr_t addr;
+
+	dma_mask = dma_alloc_coherent_mask(dev, flag);
+
+	flag |= __GFP_ZERO;
+again:
+	page = alloc_pages_node(dev_to_node(dev), flag, get_order(size));
+	if (!page)
+		return NULL;
+
+	addr = page_to_phys(page);
+	if (!is_buffer_dma_capable(dma_mask, addr, size)) {
+		__free_pages(page, get_order(size));
+
+		if (dma_mask < DMA_32BIT_MASK && !(flag & GFP_DMA)) {
+			flag = (flag & ~GFP_DMA32) | GFP_DMA;
+			goto again;
+		}
+
+		return NULL;
+	}
+
+	*dma_addr = addr;
+	return page_address(page);
+}
+
 /*
  * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter
  * documentation.
@@ -241,147 +273,6 @@ int dma_supported(struct device *dev, u64 mask)
 }
 EXPORT_SYMBOL(dma_supported);
 
-/* Allocate DMA memory on node near device */
-static noinline struct page *
-dma_alloc_pages(struct device *dev, gfp_t gfp, unsigned order)
-{
-	int node;
-
-	node = dev_to_node(dev);
-
-	return alloc_pages_node(node, gfp, order);
-}
-
-/*
- * Allocate memory for a coherent mapping.
- */
-void *
-dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		   gfp_t gfp)
-{
-	struct dma_mapping_ops *ops = get_dma_ops(dev);
-	void *memory = NULL;
-	struct page *page;
-	unsigned long dma_mask = 0;
-	dma_addr_t bus;
-	int noretry = 0;
-
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
-
-	if (dma_alloc_from_coherent(dev, size, dma_handle, &memory))
-		return memory;
-
-	if (!dev) {
-		dev = &fallback_dev;
-		gfp |= GFP_DMA;
-	}
-	dma_mask = dev->coherent_dma_mask;
-	if (dma_mask == 0)
-		dma_mask = (gfp & GFP_DMA) ? DMA_24BIT_MASK : DMA_32BIT_MASK;
-
-	/* Device not DMA able */
-	if (dev->dma_mask == NULL)
-		return NULL;
-
-	/* Don't invoke OOM killer or retry in lower 16MB DMA zone */
-	if (gfp & __GFP_DMA)
-		noretry = 1;
-
-#ifdef CONFIG_X86_64
-	/* Why <=? Even when the mask is smaller than 4GB it is often
-	   larger than 16MB and in this case we have a chance of
-	   finding fitting memory in the next higher zone first. If
-	   not retry with true GFP_DMA. -AK */
-	if (dma_mask <= DMA_32BIT_MASK && !(gfp & GFP_DMA)) {
-		gfp |= GFP_DMA32;
-		if (dma_mask < DMA_32BIT_MASK)
-			noretry = 1;
-	}
-#endif
-
- again:
-	page = dma_alloc_pages(dev,
-		noretry ? gfp | __GFP_NORETRY : gfp, get_order(size));
-	if (page == NULL)
-		return NULL;
-
-	{
-		int high, mmu;
-		bus = page_to_phys(page);
-		memory = page_address(page);
-		high = (bus + size) >= dma_mask;
-		mmu = high;
-		if (force_iommu && !(gfp & GFP_DMA))
-			mmu = 1;
-		else if (high) {
-			free_pages((unsigned long)memory,
-				   get_order(size));
-
-			/* Don't use the 16MB ZONE_DMA unless absolutely
-			   needed. It's better to use remapping first. */
-			if (dma_mask < DMA_32BIT_MASK && !(gfp & GFP_DMA)) {
-				gfp = (gfp & ~GFP_DMA32) | GFP_DMA;
-				goto again;
-			}
-
-			/* Let low level make its own zone decisions */
-			gfp &= ~(GFP_DMA32|GFP_DMA);
-
-			if (ops->alloc_coherent)
-				return ops->alloc_coherent(dev, size,
-							   dma_handle, gfp);
-			return NULL;
-		}
-
-		memset(memory, 0, size);
-		if (!mmu) {
-			*dma_handle = bus;
-			return memory;
-		}
-	}
-
-	if (ops->alloc_coherent) {
-		free_pages((unsigned long)memory, get_order(size));
-		gfp &= ~(GFP_DMA|GFP_DMA32);
-		return ops->alloc_coherent(dev, size, dma_handle, gfp);
-	}
-
-	if (ops->map_simple) {
-		*dma_handle = ops->map_simple(dev, virt_to_phys(memory),
-					      size,
-					      PCI_DMA_BIDIRECTIONAL);
-		if (*dma_handle != bad_dma_address)
-			return memory;
-	}
-
-	if (panic_on_overflow)
-		panic("dma_alloc_coherent: IOMMU overflow by %lu bytes\n",
-		      (unsigned long)size);
-	free_pages((unsigned long)memory, get_order(size));
-	return NULL;
-}
-EXPORT_SYMBOL(dma_alloc_coherent);
-
-/*
- * Unmap coherent memory.
- * The caller must ensure that the device has finished accessing the mapping.
- */
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t bus)
-{
-	struct dma_mapping_ops *ops = get_dma_ops(dev);
-
-	int order = get_order(size);
-	WARN_ON(irqs_disabled());	/* for portability */
-	if (dma_release_from_coherent(dev, order, vaddr))
-		return;
-	if (ops->unmap_single)
-		ops->unmap_single(dev, bus, size, 0);
-	free_pages((unsigned long)vaddr, order);
-}
-EXPORT_SYMBOL(dma_free_coherent);
-
 static int __init pci_iommu_init(void)
 {
 	calgary_iommu_init();
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index 49285f8fd4d..145f1c83369 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -27,8 +27,8 @@
 #include <linux/scatterlist.h>
 #include <linux/iommu-helper.h>
 #include <linux/sysdev.h>
+#include <linux/io.h>
 #include <asm/atomic.h>
-#include <asm/io.h>
 #include <asm/mtrr.h>
 #include <asm/pgtable.h>
 #include <asm/proto.h>
@@ -80,9 +80,10 @@ AGPEXTERN int agp_memory_reserved;
 AGPEXTERN __u32 *agp_gatt_table;
 
 static unsigned long next_bit;  /* protected by iommu_bitmap_lock */
-static int need_flush;		/* global flush state. set for each gart wrap */
+static bool need_flush;		/* global flush state. set for each gart wrap */
 
-static unsigned long alloc_iommu(struct device *dev, int size)
+static unsigned long alloc_iommu(struct device *dev, int size,
+				 unsigned long align_mask)
 {
 	unsigned long offset, flags;
 	unsigned long boundary_size;
@@ -90,26 +91,27 @@ static unsigned long alloc_iommu(struct device *dev, int size)
 
 	base_index = ALIGN(iommu_bus_base & dma_get_seg_boundary(dev),
 			   PAGE_SIZE) >> PAGE_SHIFT;
-	boundary_size = ALIGN(dma_get_seg_boundary(dev) + 1,
+	boundary_size = ALIGN((unsigned long long)dma_get_seg_boundary(dev) + 1,
 			      PAGE_SIZE) >> PAGE_SHIFT;
 
 	spin_lock_irqsave(&iommu_bitmap_lock, flags);
 	offset = iommu_area_alloc(iommu_gart_bitmap, iommu_pages, next_bit,
-				  size, base_index, boundary_size, 0);
+				  size, base_index, boundary_size, align_mask);
 	if (offset == -1) {
-		need_flush = 1;
+		need_flush = true;
 		offset = iommu_area_alloc(iommu_gart_bitmap, iommu_pages, 0,
-					  size, base_index, boundary_size, 0);
+					  size, base_index, boundary_size,
+					  align_mask);
 	}
 	if (offset != -1) {
 		next_bit = offset+size;
 		if (next_bit >= iommu_pages) {
 			next_bit = 0;
-			need_flush = 1;
+			need_flush = true;
 		}
 	}
 	if (iommu_fullflush)
-		need_flush = 1;
+		need_flush = true;
 	spin_unlock_irqrestore(&iommu_bitmap_lock, flags);
 
 	return offset;
@@ -134,7 +136,7 @@ static void flush_gart(void)
 	spin_lock_irqsave(&iommu_bitmap_lock, flags);
 	if (need_flush) {
 		k8_flush_garts();
-		need_flush = 0;
+		need_flush = false;
 	}
 	spin_unlock_irqrestore(&iommu_bitmap_lock, flags);
 }
@@ -173,7 +175,8 @@ static void dump_leak(void)
 	       iommu_leak_pages);
 	for (i = 0; i < iommu_leak_pages; i += 2) {
 		printk(KERN_DEBUG "%lu: ", iommu_pages-i);
-		printk_address((unsigned long) iommu_leak_tab[iommu_pages-i], 0);
+		printk_address((unsigned long) iommu_leak_tab[iommu_pages-i],
+				0);
 		printk(KERN_CONT "%c", (i+1)%2 == 0 ? '\n' : ' ');
 	}
 	printk(KERN_DEBUG "\n");
@@ -212,34 +215,24 @@ static void iommu_full(struct device *dev, size_t size, int dir)
 static inline int
 need_iommu(struct device *dev, unsigned long addr, size_t size)
 {
-	u64 mask = *dev->dma_mask;
-	int high = addr + size > mask;
-	int mmu = high;
-
-	if (force_iommu)
-		mmu = 1;
-
-	return mmu;
+	return force_iommu ||
+		!is_buffer_dma_capable(*dev->dma_mask, addr, size);
 }
 
 static inline int
 nonforced_iommu(struct device *dev, unsigned long addr, size_t size)
 {
-	u64 mask = *dev->dma_mask;
-	int high = addr + size > mask;
-	int mmu = high;
-
-	return mmu;
+	return !is_buffer_dma_capable(*dev->dma_mask, addr, size);
 }
 
 /* Map a single continuous physical area into the IOMMU.
  * Caller needs to check if the iommu is needed and flush.
  */
 static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem,
-				size_t size, int dir)
+				size_t size, int dir, unsigned long align_mask)
 {
 	unsigned long npages = iommu_num_pages(phys_mem, size);
-	unsigned long iommu_page = alloc_iommu(dev, npages);
+	unsigned long iommu_page = alloc_iommu(dev, npages, align_mask);
 	int i;
 
 	if (iommu_page == -1) {
@@ -259,16 +252,6 @@ static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem,
 	return iommu_bus_base + iommu_page*PAGE_SIZE + (phys_mem & ~PAGE_MASK);
 }
 
-static dma_addr_t
-gart_map_simple(struct device *dev, phys_addr_t paddr, size_t size, int dir)
-{
-	dma_addr_t map = dma_map_area(dev, paddr, size, dir);
-
-	flush_gart();
-
-	return map;
-}
-
 /* Map a single area into the IOMMU */
 static dma_addr_t
 gart_map_single(struct device *dev, phys_addr_t paddr, size_t size, int dir)
@@ -276,12 +259,13 @@ gart_map_single(struct device *dev, phys_addr_t paddr, size_t size, int dir)
 	unsigned long bus;
 
 	if (!dev)
-		dev = &fallback_dev;
+		dev = &x86_dma_fallback_dev;
 
 	if (!need_iommu(dev, paddr, size))
 		return paddr;
 
-	bus = gart_map_simple(dev, paddr, size, dir);
+	bus = dma_map_area(dev, paddr, size, dir, 0);
+	flush_gart();
 
 	return bus;
 }
@@ -340,7 +324,7 @@ static int dma_map_sg_nonforce(struct device *dev, struct scatterlist *sg,
 		unsigned long addr = sg_phys(s);
 
 		if (nonforced_iommu(dev, addr, s->length)) {
-			addr = dma_map_area(dev, addr, s->length, dir);
+			addr = dma_map_area(dev, addr, s->length, dir, 0);
 			if (addr == bad_dma_address) {
 				if (i > 0)
 					gart_unmap_sg(dev, sg, i, dir);
@@ -362,7 +346,7 @@ static int __dma_map_cont(struct device *dev, struct scatterlist *start,
 			  int nelems, struct scatterlist *sout,
 			  unsigned long pages)
 {
-	unsigned long iommu_start = alloc_iommu(dev, pages);
+	unsigned long iommu_start = alloc_iommu(dev, pages, 0);
 	unsigned long iommu_page = iommu_start;
 	struct scatterlist *s;
 	int i;
@@ -427,7 +411,7 @@ gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir)
 		return 0;
 
 	if (!dev)
-		dev = &fallback_dev;
+		dev = &x86_dma_fallback_dev;
 
 	out = 0;
 	start = 0;
@@ -499,6 +483,46 @@ error:
 	return 0;
 }
 
+/* allocate and map a coherent mapping */
+static void *
+gart_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr,
+		    gfp_t flag)
+{
+	dma_addr_t paddr;
+	unsigned long align_mask;
+	struct page *page;
+
+	if (force_iommu && !(flag & GFP_DMA)) {
+		flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
+		page = alloc_pages(flag | __GFP_ZERO, get_order(size));
+		if (!page)
+			return NULL;
+
+		align_mask = (1UL << get_order(size)) - 1;
+		paddr = dma_map_area(dev, page_to_phys(page), size,
+				     DMA_BIDIRECTIONAL, align_mask);
+
+		flush_gart();
+		if (paddr != bad_dma_address) {
+			*dma_addr = paddr;
+			return page_address(page);
+		}
+		__free_pages(page, get_order(size));
+	} else
+		return dma_generic_alloc_coherent(dev, size, dma_addr, flag);
+
+	return NULL;
+}
+
+/* free a coherent mapping */
+static void
+gart_free_coherent(struct device *dev, size_t size, void *vaddr,
+		   dma_addr_t dma_addr)
+{
+	gart_unmap_single(dev, dma_addr, size, DMA_BIDIRECTIONAL);
+	free_pages((unsigned long)vaddr, get_order(size));
+}
+
 static int no_agp;
 
 static __init unsigned long check_iommu_size(unsigned long aper, u64 aper_size)
@@ -626,7 +650,6 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 	struct pci_dev *dev;
 	void *gatt;
 	int i, error;
-	unsigned long start_pfn, end_pfn;
 
 	printk(KERN_INFO "PCI-DMA: Disabling AGP.\n");
 	aper_size = aper_base = info->aper_size = 0;
@@ -650,13 +673,13 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 	info->aper_size = aper_size >> 20;
 
 	gatt_size = (aper_size >> PAGE_SHIFT) * sizeof(u32);
-	gatt = (void *)__get_free_pages(GFP_KERNEL, get_order(gatt_size));
+	gatt = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
+					get_order(gatt_size));
 	if (!gatt)
 		panic("Cannot allocate GATT table");
 	if (set_memory_uc((unsigned long)gatt, gatt_size >> PAGE_SHIFT))
 		panic("Could not set GART PTEs to uncacheable pages");
 
-	memset(gatt, 0, gatt_size);
 	agp_gatt_table = gatt;
 
 	enable_gart_translations();
@@ -665,19 +688,14 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 	if (!error)
 		error = sysdev_register(&device_gart);
 	if (error)
-		panic("Could not register gart_sysdev -- would corrupt data on next suspend");
+		panic("Could not register gart_sysdev -- "
+		      "would corrupt data on next suspend");
 
 	flush_gart();
 
 	printk(KERN_INFO "PCI-DMA: aperture base @ %x size %u KB\n",
 	       aper_base, aper_size>>10);
 
-	/* need to map that range */
-	end_pfn = (aper_base>>PAGE_SHIFT) + (aper_size>>PAGE_SHIFT);
-	if (end_pfn > max_low_pfn_mapped) {
-		start_pfn = (aper_base>>PAGE_SHIFT);
-		init_memory_mapping(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);
-	}
 	return 0;
 
  nommu:
@@ -687,20 +705,13 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 	return -1;
 }
 
-extern int agp_amd64_init(void);
-
 static struct dma_mapping_ops gart_dma_ops = {
 	.map_single			= gart_map_single,
-	.map_simple			= gart_map_simple,
 	.unmap_single			= gart_unmap_single,
-	.sync_single_for_cpu		= NULL,
-	.sync_single_for_device		= NULL,
-	.sync_single_range_for_cpu	= NULL,
-	.sync_single_range_for_device	= NULL,
-	.sync_sg_for_cpu		= NULL,
-	.sync_sg_for_device		= NULL,
 	.map_sg				= gart_map_sg,
 	.unmap_sg			= gart_unmap_sg,
+	.alloc_coherent			= gart_alloc_coherent,
+	.free_coherent			= gart_free_coherent,
 };
 
 void gart_iommu_shutdown(void)
@@ -727,7 +738,8 @@ void __init gart_iommu_init(void)
 {
 	struct agp_kern_info info;
 	unsigned long iommu_start;
-	unsigned long aper_size;
+	unsigned long aper_base, aper_size;
+	unsigned long start_pfn, end_pfn;
 	unsigned long scratch;
 	long i;
 
@@ -759,30 +771,35 @@ void __init gart_iommu_init(void)
 	    (no_agp && init_k8_gatt(&info) < 0)) {
 		if (max_pfn > MAX_DMA32_PFN) {
 			printk(KERN_WARNING "More than 4GB of memory "
-			       	          "but GART IOMMU not available.\n"
-			       KERN_WARNING "falling back to iommu=soft.\n");
+			       "but GART IOMMU not available.\n");
+			printk(KERN_WARNING "falling back to iommu=soft.\n");
 		}
 		return;
 	}
 
+	/* need to map that range */
+	aper_size = info.aper_size << 20;
+	aper_base = info.aper_base;
+	end_pfn = (aper_base>>PAGE_SHIFT) + (aper_size>>PAGE_SHIFT);
+	if (end_pfn > max_low_pfn_mapped) {
+		start_pfn = (aper_base>>PAGE_SHIFT);
+		init_memory_mapping(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);
+	}
+
 	printk(KERN_INFO "PCI-DMA: using GART IOMMU.\n");
-	aper_size = info.aper_size * 1024 * 1024;
 	iommu_size = check_iommu_size(info.aper_base, aper_size);
 	iommu_pages = iommu_size >> PAGE_SHIFT;
 
-	iommu_gart_bitmap = (void *) __get_free_pages(GFP_KERNEL,
+	iommu_gart_bitmap = (void *) __get_free_pages(GFP_KERNEL | __GFP_ZERO,
 						      get_order(iommu_pages/8));
 	if (!iommu_gart_bitmap)
 		panic("Cannot allocate iommu bitmap\n");
-	memset(iommu_gart_bitmap, 0, iommu_pages/8);
 
 #ifdef CONFIG_IOMMU_LEAK
 	if (leak_trace) {
-		iommu_leak_tab = (void *)__get_free_pages(GFP_KERNEL,
+		iommu_leak_tab = (void *)__get_free_pages(GFP_KERNEL|__GFP_ZERO,
 				  get_order(iommu_pages*sizeof(void *)));
-		if (iommu_leak_tab)
-			memset(iommu_leak_tab, 0, iommu_pages * 8);
-		else
+		if (!iommu_leak_tab)
 			printk(KERN_DEBUG
 			       "PCI-DMA: Cannot allocate leak trace area\n");
 	}
@@ -792,7 +809,7 @@ void __init gart_iommu_init(void)
 	 * Out of IOMMU space handling.
 	 * Reserve some invalid pages at the beginning of the GART.
 	 */
-	set_bit_string(iommu_gart_bitmap, 0, EMERGENCY_PAGES);
+	iommu_area_reserve(iommu_gart_bitmap, 0, EMERGENCY_PAGES);
 
 	agp_memory_reserved = iommu_size;
 	printk(KERN_INFO
@@ -850,7 +867,8 @@ void __init gart_parse_options(char *p)
 	if (!strncmp(p, "leak", 4)) {
 		leak_trace = 1;
 		p += 4;
-		if (*p == '=') ++p;
+		if (*p == '=')
+			++p;
 		if (isdigit(*p) && get_option(&p, &arg))
 			iommu_leak_pages = arg;
 	}
diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c
index 3f91f71cdc3..c70ab5a5d4c 100644
--- a/arch/x86/kernel/pci-nommu.c
+++ b/arch/x86/kernel/pci-nommu.c
@@ -14,7 +14,7 @@
 static int
 check_addr(char *name, struct device *hwdev, dma_addr_t bus, size_t size)
 {
-	if (hwdev && bus + size > *hwdev->dma_mask) {
+	if (hwdev && !is_buffer_dma_capable(*hwdev->dma_mask, bus, size)) {
 		if (*hwdev->dma_mask >= DMA_32BIT_MASK)
 			printk(KERN_ERR
 			    "nommu_%s: overflow %Lx+%zu of device mask %Lx\n",
@@ -72,7 +72,15 @@ static int nommu_map_sg(struct device *hwdev, struct scatterlist *sg,
 	return nents;
 }
 
+static void nommu_free_coherent(struct device *dev, size_t size, void *vaddr,
+				dma_addr_t dma_addr)
+{
+	free_pages((unsigned long)vaddr, get_order(size));
+}
+
 struct dma_mapping_ops nommu_dma_ops = {
+	.alloc_coherent = dma_generic_alloc_coherent,
+	.free_coherent = nommu_free_coherent,
 	.map_single = nommu_map_single,
 	.map_sg = nommu_map_sg,
 	.is_phys = 1,
diff --git a/arch/x86/kernel/pcspeaker.c b/arch/x86/kernel/pcspeaker.c
index bc1f2d3ea27..a311ffcaad1 100644
--- a/arch/x86/kernel/pcspeaker.c
+++ b/arch/x86/kernel/pcspeaker.c
@@ -1,20 +1,13 @@
 #include <linux/platform_device.h>
-#include <linux/errno.h>
+#include <linux/err.h>
 #include <linux/init.h>
 
 static __init int add_pcspkr(void)
 {
 	struct platform_device *pd;
-	int ret;
 
-	pd = platform_device_alloc("pcspkr", -1);
-	if (!pd)
-		return -ENOMEM;
+	pd = platform_device_register_simple("pcspkr", -1, NULL, 0);
 
-	ret = platform_device_add(pd);
-	if (ret)
-		platform_device_put(pd);
-
-	return ret;
+	return IS_ERR(pd) ? PTR_ERR(pd) : 0;
 }
 device_initcall(add_pcspkr);
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 7fc4d5b0a6a..c622772744d 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -15,7 +15,6 @@ unsigned long idle_nomwait;
 EXPORT_SYMBOL(idle_nomwait);
 
 struct kmem_cache *task_xstate_cachep;
-static int force_mwait __cpuinitdata;
 
 int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
 {
@@ -185,7 +184,8 @@ static void mwait_idle(void)
 static void poll_idle(void)
 {
 	local_irq_enable();
-	cpu_relax();
+	while (!need_resched())
+		cpu_relax();
 }
 
 /*
@@ -246,6 +246,14 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c)
 	return 1;
 }
 
+static cpumask_t c1e_mask = CPU_MASK_NONE;
+static int c1e_detected;
+
+void c1e_remove_cpu(int cpu)
+{
+	cpu_clear(cpu, c1e_mask);
+}
+
 /*
  * C1E aware idle routine. We check for C1E active in the interrupt
  * pending message MSR. If we detect C1E, then we handle it the same
@@ -253,9 +261,6 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c)
  */
 static void c1e_idle(void)
 {
-	static cpumask_t c1e_mask = CPU_MASK_NONE;
-	static int c1e_detected;
-
 	if (need_resched())
 		return;
 
@@ -265,8 +270,10 @@ static void c1e_idle(void)
 		rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);
 		if (lo & K8_INTP_C1E_ACTIVE_MASK) {
 			c1e_detected = 1;
-			mark_tsc_unstable("TSC halt in C1E");
-			printk(KERN_INFO "System has C1E enabled\n");
+			if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC))
+				mark_tsc_unstable("TSC halt in AMD C1E");
+			printk(KERN_INFO "System has AMD C1E enabled\n");
+			set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E);
 		}
 	}
 
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 2c9abc95e02..205188db962 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -37,6 +37,7 @@
 #include <linux/tick.h>
 #include <linux/percpu.h>
 #include <linux/prctl.h>
+#include <linux/dmi.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -55,6 +56,7 @@
 #include <asm/tlbflush.h>
 #include <asm/cpu.h>
 #include <asm/kdebug.h>
+#include <asm/idle.h>
 #include <asm/syscalls.h>
 #include <asm/smp.h>
 
@@ -90,6 +92,7 @@ static void cpu_exit_clear(void)
 	cpu_clear(cpu, cpu_callin_map);
 
 	numa_remove_cpu(cpu);
+	c1e_remove_cpu(cpu);
 }
 
 /* We don't actually take CPU down, just spin without interrupts. */
@@ -161,6 +164,7 @@ void __show_registers(struct pt_regs *regs, int all)
 	unsigned long d0, d1, d2, d3, d6, d7;
 	unsigned long sp;
 	unsigned short ss, gs;
+	const char *board;
 
 	if (user_mode_vm(regs)) {
 		sp = regs->sp;
@@ -173,11 +177,15 @@ void __show_registers(struct pt_regs *regs, int all)
 	}
 
 	printk("\n");
-	printk("Pid: %d, comm: %s %s (%s %.*s)\n",
+
+	board = dmi_get_system_info(DMI_PRODUCT_NAME);
+	if (!board)
+		board = "";
+	printk("Pid: %d, comm: %s %s (%s %.*s) %s\n",
 			task_pid_nr(current), current->comm,
 			print_tainted(), init_utsname()->release,
 			(int)strcspn(init_utsname()->version, " "),
-			init_utsname()->version);
+			init_utsname()->version, board);
 
 	printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n",
 			(u16)regs->cs, regs->ip, regs->flags,
@@ -277,6 +285,14 @@ void exit_thread(void)
 		tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET;
 		put_cpu();
 	}
+#ifdef CONFIG_X86_DS
+	/* Free any DS contexts that have not been properly released. */
+	if (unlikely(current->thread.ds_ctx)) {
+		/* we clear debugctl to make sure DS is not used. */
+		update_debugctlmsr(0);
+		ds_free(current->thread.ds_ctx);
+	}
+#endif /* CONFIG_X86_DS */
 }
 
 void flush_thread(void)
@@ -438,6 +454,35 @@ int set_tsc_mode(unsigned int val)
 	return 0;
 }
 
+#ifdef CONFIG_X86_DS
+static int update_debugctl(struct thread_struct *prev,
+			struct thread_struct *next, unsigned long debugctl)
+{
+	unsigned long ds_prev = 0;
+	unsigned long ds_next = 0;
+
+	if (prev->ds_ctx)
+		ds_prev = (unsigned long)prev->ds_ctx->ds;
+	if (next->ds_ctx)
+		ds_next = (unsigned long)next->ds_ctx->ds;
+
+	if (ds_next != ds_prev) {
+		/* we clear debugctl to make sure DS
+		 * is not in use when we change it */
+		debugctl = 0;
+		update_debugctlmsr(0);
+		wrmsr(MSR_IA32_DS_AREA, ds_next, 0);
+	}
+	return debugctl;
+}
+#else
+static int update_debugctl(struct thread_struct *prev,
+			struct thread_struct *next, unsigned long debugctl)
+{
+	return debugctl;
+}
+#endif /* CONFIG_X86_DS */
+
 static noinline void
 __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		 struct tss_struct *tss)
@@ -448,14 +493,7 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 	prev = &prev_p->thread;
 	next = &next_p->thread;
 
-	debugctl = prev->debugctlmsr;
-	if (next->ds_area_msr != prev->ds_area_msr) {
-		/* we clear debugctl to make sure DS
-		 * is not in use when we change it */
-		debugctl = 0;
-		update_debugctlmsr(0);
-		wrmsr(MSR_IA32_DS_AREA, next->ds_area_msr, 0);
-	}
+	debugctl = update_debugctl(prev, next, prev->debugctlmsr);
 
 	if (next->debugctlmsr != debugctl)
 		update_debugctlmsr(next->debugctlmsr);
@@ -479,13 +517,13 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 			hard_enable_TSC();
 	}
 
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
 	if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
 		ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
 
 	if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
 		ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
-#endif
+#endif /* CONFIG_X86_PTRACE_BTS */
 
 
 	if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 00263c9e650..2a8ccb9238b 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -37,11 +37,11 @@
 #include <linux/kdebug.h>
 #include <linux/tick.h>
 #include <linux/prctl.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
 
-#include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/system.h>
-#include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/i387.h>
 #include <asm/mmu_context.h>
@@ -89,11 +89,13 @@ void exit_idle(void)
 #ifdef CONFIG_HOTPLUG_CPU
 DECLARE_PER_CPU(int, cpu_state);
 
-#include <asm/nmi.h>
+#include <linux/nmi.h>
 /* We halt the CPU with physical CPU hotplug */
 static inline void play_dead(void)
 {
 	idle_task_exit();
+	c1e_remove_cpu(raw_smp_processor_id());
+
 	mb();
 	/* Ack it */
 	__get_cpu_var(cpu_state) = CPU_DEAD;
@@ -152,7 +154,7 @@ void cpu_idle(void)
 }
 
 /* Prints also some state that isn't saved in the pt_regs */
-void __show_regs(struct pt_regs * regs)
+void __show_regs(struct pt_regs *regs)
 {
 	unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L, fs, gs, shadowgs;
 	unsigned long d0, d1, d2, d3, d6, d7;
@@ -161,59 +163,61 @@ void __show_regs(struct pt_regs * regs)
 
 	printk("\n");
 	print_modules();
-	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
+	printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s\n",
 		current->pid, current->comm, print_tainted(),
 		init_utsname()->release,
 		(int)strcspn(init_utsname()->version, " "),
 		init_utsname()->version);
-	printk("RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip);
+	printk(KERN_INFO "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip);
 	printk_address(regs->ip, 1);
-	printk("RSP: %04lx:%016lx  EFLAGS: %08lx\n", regs->ss, regs->sp,
-		regs->flags);
-	printk("RAX: %016lx RBX: %016lx RCX: %016lx\n",
+	printk(KERN_INFO "RSP: %04lx:%016lx  EFLAGS: %08lx\n", regs->ss,
+			regs->sp, regs->flags);
+	printk(KERN_INFO "RAX: %016lx RBX: %016lx RCX: %016lx\n",
 	       regs->ax, regs->bx, regs->cx);
-	printk("RDX: %016lx RSI: %016lx RDI: %016lx\n",
+	printk(KERN_INFO "RDX: %016lx RSI: %016lx RDI: %016lx\n",
 	       regs->dx, regs->si, regs->di);
-	printk("RBP: %016lx R08: %016lx R09: %016lx\n",
+	printk(KERN_INFO "RBP: %016lx R08: %016lx R09: %016lx\n",
 	       regs->bp, regs->r8, regs->r9);
-	printk("R10: %016lx R11: %016lx R12: %016lx\n",
-	       regs->r10, regs->r11, regs->r12); 
-	printk("R13: %016lx R14: %016lx R15: %016lx\n",
-	       regs->r13, regs->r14, regs->r15); 
-
-	asm("movl %%ds,%0" : "=r" (ds)); 
-	asm("movl %%cs,%0" : "=r" (cs)); 
-	asm("movl %%es,%0" : "=r" (es)); 
+	printk(KERN_INFO "R10: %016lx R11: %016lx R12: %016lx\n",
+	       regs->r10, regs->r11, regs->r12);
+	printk(KERN_INFO "R13: %016lx R14: %016lx R15: %016lx\n",
+	       regs->r13, regs->r14, regs->r15);
+
+	asm("movl %%ds,%0" : "=r" (ds));
+	asm("movl %%cs,%0" : "=r" (cs));
+	asm("movl %%es,%0" : "=r" (es));
 	asm("movl %%fs,%0" : "=r" (fsindex));
 	asm("movl %%gs,%0" : "=r" (gsindex));
 
 	rdmsrl(MSR_FS_BASE, fs);
-	rdmsrl(MSR_GS_BASE, gs); 
-	rdmsrl(MSR_KERNEL_GS_BASE, shadowgs); 
+	rdmsrl(MSR_GS_BASE, gs);
+	rdmsrl(MSR_KERNEL_GS_BASE, shadowgs);
 
 	cr0 = read_cr0();
 	cr2 = read_cr2();
 	cr3 = read_cr3();
 	cr4 = read_cr4();
 
-	printk("FS:  %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n", 
-	       fs,fsindex,gs,gsindex,shadowgs); 
-	printk("CS:  %04x DS: %04x ES: %04x CR0: %016lx\n", cs, ds, es, cr0); 
-	printk("CR2: %016lx CR3: %016lx CR4: %016lx\n", cr2, cr3, cr4);
+	printk(KERN_INFO "FS:  %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n",
+	       fs, fsindex, gs, gsindex, shadowgs);
+	printk(KERN_INFO "CS:  %04x DS: %04x ES: %04x CR0: %016lx\n", cs, ds,
+			es, cr0);
+	printk(KERN_INFO "CR2: %016lx CR3: %016lx CR4: %016lx\n", cr2, cr3,
+			cr4);
 
 	get_debugreg(d0, 0);
 	get_debugreg(d1, 1);
 	get_debugreg(d2, 2);
-	printk("DR0: %016lx DR1: %016lx DR2: %016lx\n", d0, d1, d2);
+	printk(KERN_INFO "DR0: %016lx DR1: %016lx DR2: %016lx\n", d0, d1, d2);
 	get_debugreg(d3, 3);
 	get_debugreg(d6, 6);
 	get_debugreg(d7, 7);
-	printk("DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7);
+	printk(KERN_INFO "DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7);
 }
 
 void show_regs(struct pt_regs *regs)
 {
-	printk("CPU %d:", smp_processor_id());
+	printk(KERN_INFO "CPU %d:", smp_processor_id());
 	__show_regs(regs);
 	show_trace(NULL, regs, (void *)(regs + 1), regs->bp);
 }
@@ -239,6 +243,14 @@ void exit_thread(void)
 		t->io_bitmap_max = 0;
 		put_cpu();
 	}
+#ifdef CONFIG_X86_DS
+	/* Free any DS contexts that have not been properly released. */
+	if (unlikely(t->ds_ctx)) {
+		/* we clear debugctl to make sure DS is not used. */
+		update_debugctlmsr(0);
+		ds_free(t->ds_ctx);
+	}
+#endif /* CONFIG_X86_DS */
 }
 
 void flush_thread(void)
@@ -314,10 +326,10 @@ void prepare_to_copy(struct task_struct *tsk)
 
 int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 		unsigned long unused,
-	struct task_struct * p, struct pt_regs * regs)
+	struct task_struct *p, struct pt_regs *regs)
 {
 	int err;
-	struct pt_regs * childregs;
+	struct pt_regs *childregs;
 	struct task_struct *me = current;
 
 	childregs = ((struct pt_regs *)
@@ -362,10 +374,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 		if (test_thread_flag(TIF_IA32))
 			err = do_set_thread_area(p, -1,
 				(struct user_desc __user *)childregs->si, 0);
-		else 			
-#endif	 
-			err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8); 
-		if (err) 
+		else
+#endif
+			err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8);
+		if (err)
 			goto out;
 	}
 	err = 0;
@@ -472,13 +484,27 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 	next = &next_p->thread;
 
 	debugctl = prev->debugctlmsr;
-	if (next->ds_area_msr != prev->ds_area_msr) {
-		/* we clear debugctl to make sure DS
-		 * is not in use when we change it */
-		debugctl = 0;
-		update_debugctlmsr(0);
-		wrmsrl(MSR_IA32_DS_AREA, next->ds_area_msr);
+
+#ifdef CONFIG_X86_DS
+	{
+		unsigned long ds_prev = 0, ds_next = 0;
+
+		if (prev->ds_ctx)
+			ds_prev = (unsigned long)prev->ds_ctx->ds;
+		if (next->ds_ctx)
+			ds_next = (unsigned long)next->ds_ctx->ds;
+
+		if (ds_next != ds_prev) {
+			/*
+			 * We clear debugctl to make sure DS
+			 * is not in use when we change it:
+			 */
+			debugctl = 0;
+			update_debugctlmsr(0);
+			wrmsrl(MSR_IA32_DS_AREA, ds_next);
+		}
 	}
+#endif /* CONFIG_X86_DS */
 
 	if (next->debugctlmsr != debugctl)
 		update_debugctlmsr(next->debugctlmsr);
@@ -516,13 +542,13 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 		memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
 	}
 
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
 	if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
 		ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
 
 	if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
 		ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
-#endif
+#endif /* CONFIG_X86_PTRACE_BTS */
 }
 
 /*
@@ -544,7 +570,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	unsigned fsindex, gsindex;
 
 	/* we're going to use this soon, after a few expensive things */
-	if (next_p->fpu_counter>5)
+	if (next_p->fpu_counter > 5)
 		prefetch(next->xstate);
 
 	/*
@@ -552,13 +578,13 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	 */
 	load_sp0(tss, next);
 
-	/* 
+	/*
 	 * Switch DS and ES.
 	 * This won't pick up thread selector changes, but I guess that is ok.
 	 */
 	savesegment(es, prev->es);
 	if (unlikely(next->es | prev->es))
-		loadsegment(es, next->es); 
+		loadsegment(es, next->es);
 
 	savesegment(ds, prev->ds);
 	if (unlikely(next->ds | prev->ds))
@@ -584,7 +610,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	 */
 	arch_leave_lazy_cpu_mode();
 
-	/* 
+	/*
 	 * Switch FS and GS.
 	 *
 	 * Segment register != 0 always requires a reload.  Also
@@ -593,13 +619,13 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	 */
 	if (unlikely(fsindex | next->fsindex | prev->fs)) {
 		loadsegment(fs, next->fsindex);
-		/* 
+		/*
 		 * Check if the user used a selector != 0; if yes
 		 *  clear 64bit base, since overloaded base is always
 		 *  mapped to the Null selector
 		 */
 		if (fsindex)
-			prev->fs = 0;				
+			prev->fs = 0;
 	}
 	/* when next process has a 64bit base use it */
 	if (next->fs)
@@ -609,7 +635,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	if (unlikely(gsindex | next->gsindex | prev->gs)) {
 		load_gs_index(next->gsindex);
 		if (gsindex)
-			prev->gs = 0;				
+			prev->gs = 0;
 	}
 	if (next->gs)
 		wrmsrl(MSR_KERNEL_GS_BASE, next->gs);
@@ -618,12 +644,12 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/* Must be after DS reload */
 	unlazy_fpu(prev_p);
 
-	/* 
+	/*
 	 * Switch the PDA and FPU contexts.
 	 */
 	prev->usersp = read_pda(oldrsp);
 	write_pda(oldrsp, next->usersp);
-	write_pda(pcurrent, next_p); 
+	write_pda(pcurrent, next_p);
 
 	write_pda(kernelstack,
 		  (unsigned long)task_stack_page(next_p) +
@@ -664,7 +690,7 @@ long sys_execve(char __user *name, char __user * __user *argv,
 		char __user * __user *envp, struct pt_regs *regs)
 {
 	long error;
-	char * filename;
+	char *filename;
 
 	filename = getname(name);
 	error = PTR_ERR(filename);
@@ -722,55 +748,55 @@ asmlinkage long sys_vfork(struct pt_regs *regs)
 unsigned long get_wchan(struct task_struct *p)
 {
 	unsigned long stack;
-	u64 fp,ip;
+	u64 fp, ip;
 	int count = 0;
 
-	if (!p || p == current || p->state==TASK_RUNNING)
-		return 0; 
+	if (!p || p == current || p->state == TASK_RUNNING)
+		return 0;
 	stack = (unsigned long)task_stack_page(p);
 	if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE)
 		return 0;
 	fp = *(u64 *)(p->thread.sp);
-	do { 
+	do {
 		if (fp < (unsigned long)stack ||
 		    fp > (unsigned long)stack+THREAD_SIZE)
-			return 0; 
+			return 0;
 		ip = *(u64 *)(fp+8);
 		if (!in_sched_functions(ip))
 			return ip;
-		fp = *(u64 *)fp; 
-	} while (count++ < 16); 
+		fp = *(u64 *)fp;
+	} while (count++ < 16);
 	return 0;
 }
 
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
-{ 
-	int ret = 0; 
+{
+	int ret = 0;
 	int doit = task == current;
 	int cpu;
 
-	switch (code) { 
+	switch (code) {
 	case ARCH_SET_GS:
 		if (addr >= TASK_SIZE_OF(task))
-			return -EPERM; 
+			return -EPERM;
 		cpu = get_cpu();
-		/* handle small bases via the GDT because that's faster to 
+		/* handle small bases via the GDT because that's faster to
 		   switch. */
-		if (addr <= 0xffffffff) {  
-			set_32bit_tls(task, GS_TLS, addr); 
-			if (doit) { 
+		if (addr <= 0xffffffff) {
+			set_32bit_tls(task, GS_TLS, addr);
+			if (doit) {
 				load_TLS(&task->thread, cpu);
-				load_gs_index(GS_TLS_SEL); 
+				load_gs_index(GS_TLS_SEL);
 			}
-			task->thread.gsindex = GS_TLS_SEL; 
+			task->thread.gsindex = GS_TLS_SEL;
 			task->thread.gs = 0;
-		} else { 
+		} else {
 			task->thread.gsindex = 0;
 			task->thread.gs = addr;
 			if (doit) {
 				load_gs_index(0);
 				ret = checking_wrmsrl(MSR_KERNEL_GS_BASE, addr);
-			} 
+			}
 		}
 		put_cpu();
 		break;
@@ -824,8 +850,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
 				rdmsrl(MSR_KERNEL_GS_BASE, base);
 			else
 				base = task->thread.gs;
-		}
-		else
+		} else
 			base = task->thread.gs;
 		ret = put_user(base, (unsigned long __user *)addr);
 		break;
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index fc3e8dcd9da..e375b658efc 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -14,6 +14,7 @@
 #include <linux/errno.h>
 #include <linux/ptrace.h>
 #include <linux/regset.h>
+#include <linux/tracehook.h>
 #include <linux/user.h>
 #include <linux/elf.h>
 #include <linux/security.h>
@@ -554,45 +555,115 @@ static int ptrace_set_debugreg(struct task_struct *child,
 	return 0;
 }
 
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
+/*
+ * The configuration for a particular BTS hardware implementation.
+ */
+struct bts_configuration {
+	/* the size of a BTS record in bytes; at most BTS_MAX_RECORD_SIZE */
+	unsigned char  sizeof_bts;
+	/* the size of a field in the BTS record in bytes */
+	unsigned char  sizeof_field;
+	/* a bitmask to enable/disable BTS in DEBUGCTL MSR */
+	unsigned long debugctl_mask;
+};
+static struct bts_configuration bts_cfg;
+
+#define BTS_MAX_RECORD_SIZE (8 * 3)
+
+
+/*
+ * Branch Trace Store (BTS) uses the following format. Different
+ * architectures vary in the size of those fields.
+ * - source linear address
+ * - destination linear address
+ * - flags
+ *
+ * Later architectures use 64bit pointers throughout, whereas earlier
+ * architectures use 32bit pointers in 32bit mode.
+ *
+ * We compute the base address for the first 8 fields based on:
+ * - the field size stored in the DS configuration
+ * - the relative field position
+ *
+ * In order to store additional information in the BTS buffer, we use
+ * a special source address to indicate that the record requires
+ * special interpretation.
+ *
+ * Netburst indicated via a bit in the flags field whether the branch
+ * was predicted; this is ignored.
+ */
+
+enum bts_field {
+	bts_from = 0,
+	bts_to,
+	bts_flags,
+
+	bts_escape = (unsigned long)-1,
+	bts_qual = bts_to,
+	bts_jiffies = bts_flags
+};
+
+static inline unsigned long bts_get(const char *base, enum bts_field field)
+{
+	base += (bts_cfg.sizeof_field * field);
+	return *(unsigned long *)base;
+}
 
-static int ptrace_bts_get_size(struct task_struct *child)
+static inline void bts_set(char *base, enum bts_field field, unsigned long val)
 {
-	if (!child->thread.ds_area_msr)
-		return -ENXIO;
+	base += (bts_cfg.sizeof_field * field);;
+	(*(unsigned long *)base) = val;
+}
 
-	return ds_get_bts_index((void *)child->thread.ds_area_msr);
+/*
+ * Translate a BTS record from the raw format into the bts_struct format
+ *
+ * out (out): bts_struct interpretation
+ * raw: raw BTS record
+ */
+static void ptrace_bts_translate_record(struct bts_struct *out, const void *raw)
+{
+	memset(out, 0, sizeof(*out));
+	if (bts_get(raw, bts_from) == bts_escape) {
+		out->qualifier       = bts_get(raw, bts_qual);
+		out->variant.jiffies = bts_get(raw, bts_jiffies);
+	} else {
+		out->qualifier = BTS_BRANCH;
+		out->variant.lbr.from_ip = bts_get(raw, bts_from);
+		out->variant.lbr.to_ip   = bts_get(raw, bts_to);
+	}
 }
 
-static int ptrace_bts_read_record(struct task_struct *child,
-				  long index,
+static int ptrace_bts_read_record(struct task_struct *child, size_t index,
 				  struct bts_struct __user *out)
 {
 	struct bts_struct ret;
-	int retval;
-	int bts_end;
-	int bts_index;
-
-	if (!child->thread.ds_area_msr)
-		return -ENXIO;
+	const void *bts_record;
+	size_t bts_index, bts_end;
+	int error;
 
-	if (index < 0)
-		return -EINVAL;
+	error = ds_get_bts_end(child, &bts_end);
+	if (error < 0)
+		return error;
 
-	bts_end = ds_get_bts_end((void *)child->thread.ds_area_msr);
 	if (bts_end <= index)
 		return -EINVAL;
 
+	error = ds_get_bts_index(child, &bts_index);
+	if (error < 0)
+		return error;
+
 	/* translate the ptrace bts index into the ds bts index */
-	bts_index = ds_get_bts_index((void *)child->thread.ds_area_msr);
-	bts_index -= (index + 1);
-	if (bts_index < 0)
-		bts_index += bts_end;
+	bts_index += bts_end - (index + 1);
+	if (bts_end <= bts_index)
+		bts_index -= bts_end;
 
-	retval = ds_read_bts((void *)child->thread.ds_area_msr,
-			     bts_index, &ret);
-	if (retval < 0)
-		return retval;
+	error = ds_access_bts(child, bts_index, &bts_record);
+	if (error < 0)
+		return error;
+
+	ptrace_bts_translate_record(&ret, bts_record);
 
 	if (copy_to_user(out, &ret, sizeof(ret)))
 		return -EFAULT;
@@ -600,101 +671,106 @@ static int ptrace_bts_read_record(struct task_struct *child,
 	return sizeof(ret);
 }
 
-static int ptrace_bts_clear(struct task_struct *child)
-{
-	if (!child->thread.ds_area_msr)
-		return -ENXIO;
-
-	return ds_clear((void *)child->thread.ds_area_msr);
-}
-
 static int ptrace_bts_drain(struct task_struct *child,
 			    long size,
 			    struct bts_struct __user *out)
 {
-	int end, i;
-	void *ds = (void *)child->thread.ds_area_msr;
-
-	if (!ds)
-		return -ENXIO;
+	struct bts_struct ret;
+	const unsigned char *raw;
+	size_t end, i;
+	int error;
 
-	end = ds_get_bts_index(ds);
-	if (end <= 0)
-		return end;
+	error = ds_get_bts_index(child, &end);
+	if (error < 0)
+		return error;
 
 	if (size < (end * sizeof(struct bts_struct)))
 		return -EIO;
 
-	for (i = 0; i < end; i++, out++) {
-		struct bts_struct ret;
-		int retval;
+	error = ds_access_bts(child, 0, (const void **)&raw);
+	if (error < 0)
+		return error;
 
-		retval = ds_read_bts(ds, i, &ret);
-		if (retval < 0)
-			return retval;
+	for (i = 0; i < end; i++, out++, raw += bts_cfg.sizeof_bts) {
+		ptrace_bts_translate_record(&ret, raw);
 
 		if (copy_to_user(out, &ret, sizeof(ret)))
 			return -EFAULT;
 	}
 
-	ds_clear(ds);
+	error = ds_clear_bts(child);
+	if (error < 0)
+		return error;
 
 	return end;
 }
 
+static void ptrace_bts_ovfl(struct task_struct *child)
+{
+	send_sig(child->thread.bts_ovfl_signal, child, 0);
+}
+
 static int ptrace_bts_config(struct task_struct *child,
 			     long cfg_size,
 			     const struct ptrace_bts_config __user *ucfg)
 {
 	struct ptrace_bts_config cfg;
-	int bts_size, ret = 0;
-	void *ds;
+	int error = 0;
+
+	error = -EOPNOTSUPP;
+	if (!bts_cfg.sizeof_bts)
+		goto errout;
 
+	error = -EIO;
 	if (cfg_size < sizeof(cfg))
-		return -EIO;
+		goto errout;
 
+	error = -EFAULT;
 	if (copy_from_user(&cfg, ucfg, sizeof(cfg)))
-		return -EFAULT;
+		goto errout;
 
-	if ((int)cfg.size < 0)
-		return -EINVAL;
+	error = -EINVAL;
+	if ((cfg.flags & PTRACE_BTS_O_SIGNAL) &&
+	    !(cfg.flags & PTRACE_BTS_O_ALLOC))
+		goto errout;
 
-	bts_size = 0;
-	ds = (void *)child->thread.ds_area_msr;
-	if (ds) {
-		bts_size = ds_get_bts_size(ds);
-		if (bts_size < 0)
-			return bts_size;
-	}
-	cfg.size = PAGE_ALIGN(cfg.size);
+	if (cfg.flags & PTRACE_BTS_O_ALLOC) {
+		ds_ovfl_callback_t ovfl = NULL;
+		unsigned int sig = 0;
+
+		/* we ignore the error in case we were not tracing child */
+		(void)ds_release_bts(child);
 
-	if (bts_size != cfg.size) {
-		ret = ptrace_bts_realloc(child, cfg.size,
-					 cfg.flags & PTRACE_BTS_O_CUT_SIZE);
-		if (ret < 0)
+		if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
+			if (!cfg.signal)
+				goto errout;
+
+			sig  = cfg.signal;
+			ovfl = ptrace_bts_ovfl;
+		}
+
+		error = ds_request_bts(child, /* base = */ NULL, cfg.size, ovfl);
+		if (error < 0)
 			goto errout;
 
-		ds = (void *)child->thread.ds_area_msr;
+		child->thread.bts_ovfl_signal = sig;
 	}
 
-	if (cfg.flags & PTRACE_BTS_O_SIGNAL)
-		ret = ds_set_overflow(ds, DS_O_SIGNAL);
-	else
-		ret = ds_set_overflow(ds, DS_O_WRAP);
-	if (ret < 0)
+	error = -EINVAL;
+	if (!child->thread.ds_ctx && cfg.flags)
 		goto errout;
 
 	if (cfg.flags & PTRACE_BTS_O_TRACE)
-		child->thread.debugctlmsr |= ds_debugctl_mask();
+		child->thread.debugctlmsr |= bts_cfg.debugctl_mask;
 	else
-		child->thread.debugctlmsr &= ~ds_debugctl_mask();
+		child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
 
 	if (cfg.flags & PTRACE_BTS_O_SCHED)
 		set_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
 	else
 		clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
 
-	ret = sizeof(cfg);
+	error = sizeof(cfg);
 
 out:
 	if (child->thread.debugctlmsr)
@@ -702,10 +778,10 @@ out:
 	else
 		clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
 
-	return ret;
+	return error;
 
 errout:
-	child->thread.debugctlmsr &= ~ds_debugctl_mask();
+	child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
 	clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
 	goto out;
 }
@@ -714,29 +790,40 @@ static int ptrace_bts_status(struct task_struct *child,
 			     long cfg_size,
 			     struct ptrace_bts_config __user *ucfg)
 {
-	void *ds = (void *)child->thread.ds_area_msr;
 	struct ptrace_bts_config cfg;
+	size_t end;
+	const void *base, *max;
+	int error;
 
 	if (cfg_size < sizeof(cfg))
 		return -EIO;
 
-	memset(&cfg, 0, sizeof(cfg));
+	error = ds_get_bts_end(child, &end);
+	if (error < 0)
+		return error;
 
-	if (ds) {
-		cfg.size = ds_get_bts_size(ds);
+	error = ds_access_bts(child, /* index = */ 0, &base);
+	if (error < 0)
+		return error;
 
-		if (ds_get_overflow(ds) == DS_O_SIGNAL)
-			cfg.flags |= PTRACE_BTS_O_SIGNAL;
+	error = ds_access_bts(child, /* index = */ end, &max);
+	if (error < 0)
+		return error;
 
-		if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
-		    child->thread.debugctlmsr & ds_debugctl_mask())
-			cfg.flags |= PTRACE_BTS_O_TRACE;
+	memset(&cfg, 0, sizeof(cfg));
+	cfg.size = (max - base);
+	cfg.signal = child->thread.bts_ovfl_signal;
+	cfg.bts_size = sizeof(struct bts_struct);
 
-		if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
-			cfg.flags |= PTRACE_BTS_O_SCHED;
-	}
+	if (cfg.signal)
+		cfg.flags |= PTRACE_BTS_O_SIGNAL;
 
-	cfg.bts_size = sizeof(struct bts_struct);
+	if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
+	    child->thread.debugctlmsr & bts_cfg.debugctl_mask)
+		cfg.flags |= PTRACE_BTS_O_TRACE;
+
+	if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
+		cfg.flags |= PTRACE_BTS_O_SCHED;
 
 	if (copy_to_user(ucfg, &cfg, sizeof(cfg)))
 		return -EFAULT;
@@ -744,89 +831,38 @@ static int ptrace_bts_status(struct task_struct *child,
 	return sizeof(cfg);
 }
 
-
 static int ptrace_bts_write_record(struct task_struct *child,
 				   const struct bts_struct *in)
 {
-	int retval;
+	unsigned char bts_record[BTS_MAX_RECORD_SIZE];
 
-	if (!child->thread.ds_area_msr)
-		return -ENXIO;
+	BUG_ON(BTS_MAX_RECORD_SIZE < bts_cfg.sizeof_bts);
 
-	retval = ds_write_bts((void *)child->thread.ds_area_msr, in);
-	if (retval)
-		return retval;
+	memset(bts_record, 0, bts_cfg.sizeof_bts);
+	switch (in->qualifier) {
+	case BTS_INVALID:
+		break;
 
-	return sizeof(*in);
-}
+	case BTS_BRANCH:
+		bts_set(bts_record, bts_from, in->variant.lbr.from_ip);
+		bts_set(bts_record, bts_to,   in->variant.lbr.to_ip);
+		break;
 
-static int ptrace_bts_realloc(struct task_struct *child,
-			      int size, int reduce_size)
-{
-	unsigned long rlim, vm;
-	int ret, old_size;
+	case BTS_TASK_ARRIVES:
+	case BTS_TASK_DEPARTS:
+		bts_set(bts_record, bts_from,    bts_escape);
+		bts_set(bts_record, bts_qual,    in->qualifier);
+		bts_set(bts_record, bts_jiffies, in->variant.jiffies);
+		break;
 
-	if (size < 0)
+	default:
 		return -EINVAL;
-
-	old_size = ds_get_bts_size((void *)child->thread.ds_area_msr);
-	if (old_size < 0)
-		return old_size;
-
-	ret = ds_free((void **)&child->thread.ds_area_msr);
-	if (ret < 0)
-		goto out;
-
-	size >>= PAGE_SHIFT;
-	old_size >>= PAGE_SHIFT;
-
-	current->mm->total_vm  -= old_size;
-	current->mm->locked_vm -= old_size;
-
-	if (size == 0)
-		goto out;
-
-	rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT;
-	vm = current->mm->total_vm  + size;
-	if (rlim < vm) {
-		ret = -ENOMEM;
-
-		if (!reduce_size)
-			goto out;
-
-		size = rlim - current->mm->total_vm;
-		if (size <= 0)
-			goto out;
-	}
-
-	rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT;
-	vm = current->mm->locked_vm  + size;
-	if (rlim < vm) {
-		ret = -ENOMEM;
-
-		if (!reduce_size)
-			goto out;
-
-		size = rlim - current->mm->locked_vm;
-		if (size <= 0)
-			goto out;
 	}
 
-	ret = ds_allocate((void **)&child->thread.ds_area_msr,
-			  size << PAGE_SHIFT);
-	if (ret < 0)
-		goto out;
-
-	current->mm->total_vm  += size;
-	current->mm->locked_vm += size;
-
-out:
-	if (child->thread.ds_area_msr)
-		set_tsk_thread_flag(child, TIF_DS_AREA_MSR);
-	else
-		clear_tsk_thread_flag(child, TIF_DS_AREA_MSR);
-
-	return ret;
+	/* The writing task will be the switched-to task on a context
+	 * switch. It needs to write into the switched-from task's BTS
+	 * buffer. */
+	return ds_unchecked_write_bts(child, bts_record, bts_cfg.sizeof_bts);
 }
 
 void ptrace_bts_take_timestamp(struct task_struct *tsk,
@@ -839,7 +875,66 @@ void ptrace_bts_take_timestamp(struct task_struct *tsk,
 
 	ptrace_bts_write_record(tsk, &rec);
 }
-#endif /* X86_BTS */
+
+static const struct bts_configuration bts_cfg_netburst = {
+	.sizeof_bts    = sizeof(long) * 3,
+	.sizeof_field  = sizeof(long),
+	.debugctl_mask = (1<<2)|(1<<3)|(1<<5)
+};
+
+static const struct bts_configuration bts_cfg_pentium_m = {
+	.sizeof_bts    = sizeof(long) * 3,
+	.sizeof_field  = sizeof(long),
+	.debugctl_mask = (1<<6)|(1<<7)
+};
+
+static const struct bts_configuration bts_cfg_core2 = {
+	.sizeof_bts    = 8 * 3,
+	.sizeof_field  = 8,
+	.debugctl_mask = (1<<6)|(1<<7)|(1<<9)
+};
+
+static inline void bts_configure(const struct bts_configuration *cfg)
+{
+	bts_cfg = *cfg;
+}
+
+void __cpuinit ptrace_bts_init_intel(struct cpuinfo_x86 *c)
+{
+	switch (c->x86) {
+	case 0x6:
+		switch (c->x86_model) {
+		case 0xD:
+		case 0xE: /* Pentium M */
+			bts_configure(&bts_cfg_pentium_m);
+			break;
+		case 0xF: /* Core2 */
+        case 0x1C: /* Atom */
+			bts_configure(&bts_cfg_core2);
+			break;
+		default:
+			/* sorry, don't know about them */
+			break;
+		}
+		break;
+	case 0xF:
+		switch (c->x86_model) {
+		case 0x0:
+		case 0x1:
+		case 0x2: /* Netburst */
+			bts_configure(&bts_cfg_netburst);
+			break;
+		default:
+			/* sorry, don't know about them */
+			break;
+		}
+		break;
+	default:
+		/* sorry, don't know about them */
+		break;
+	}
+}
+#endif /* CONFIG_X86_PTRACE_BTS */
 
 /*
  * Called by kernel/ptrace.c when detaching..
@@ -852,15 +947,15 @@ void ptrace_disable(struct task_struct *child)
 #ifdef TIF_SYSCALL_EMU
 	clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
 #endif
-	if (child->thread.ds_area_msr) {
-#ifdef X86_BTS
-		ptrace_bts_realloc(child, 0, 0);
-#endif
-		child->thread.debugctlmsr &= ~ds_debugctl_mask();
-		if (!child->thread.debugctlmsr)
-			clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
-		clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
-	}
+#ifdef CONFIG_X86_PTRACE_BTS
+	(void)ds_release_bts(child);
+
+	child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
+	if (!child->thread.debugctlmsr)
+		clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+
+	clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
+#endif /* CONFIG_X86_PTRACE_BTS */
 }
 
 #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
@@ -980,7 +1075,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 	/*
 	 * These bits need more cooking - not enabled yet:
 	 */
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
 	case PTRACE_BTS_CONFIG:
 		ret = ptrace_bts_config
 			(child, data, (struct ptrace_bts_config __user *)addr);
@@ -992,7 +1087,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 
 	case PTRACE_BTS_SIZE:
-		ret = ptrace_bts_get_size(child);
+		ret = ds_get_bts_index(child, /* pos = */ NULL);
 		break;
 
 	case PTRACE_BTS_GET:
@@ -1001,14 +1096,14 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 
 	case PTRACE_BTS_CLEAR:
-		ret = ptrace_bts_clear(child);
+		ret = ds_clear_bts(child);
 		break;
 
 	case PTRACE_BTS_DRAIN:
 		ret = ptrace_bts_drain
 			(child, data, (struct bts_struct __user *) addr);
 		break;
-#endif
+#endif /* CONFIG_X86_PTRACE_BTS */
 
 	default:
 		ret = ptrace_request(child, request, addr, data);
@@ -1375,30 +1470,6 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code)
 	force_sig_info(SIGTRAP, &info, tsk);
 }
 
-static void syscall_trace(struct pt_regs *regs)
-{
-	if (!(current->ptrace & PT_PTRACED))
-		return;
-
-#if 0
-	printk("trace %s ip %lx sp %lx ax %d origrax %d caller %lx tiflags %x ptrace %x\n",
-	       current->comm,
-	       regs->ip, regs->sp, regs->ax, regs->orig_ax, __builtin_return_address(0),
-	       current_thread_info()->flags, current->ptrace);
-#endif
-
-	ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD)
-				? 0x80 : 0));
-	/*
-	 * this isn't the same as continuing with a signal, but it will do
-	 * for normal use.  strace only continues with a signal if the
-	 * stopping signal is not SIGTRAP.  -brl
-	 */
-	if (current->exit_code) {
-		send_sig(current->exit_code, current, 1);
-		current->exit_code = 0;
-	}
-}
 
 #ifdef CONFIG_X86_32
 # define IS_IA32	1
@@ -1432,8 +1503,9 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs)
 	if (unlikely(test_thread_flag(TIF_SYSCALL_EMU)))
 		ret = -1L;
 
-	if (ret || test_thread_flag(TIF_SYSCALL_TRACE))
-		syscall_trace(regs);
+	if ((ret || test_thread_flag(TIF_SYSCALL_TRACE)) &&
+	    tracehook_report_syscall_entry(regs))
+		ret = -1L;
 
 	if (unlikely(current->audit_context)) {
 		if (IS_IA32)
@@ -1459,7 +1531,7 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs)
 		audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
-		syscall_trace(regs);
+		tracehook_report_syscall_exit(regs, 0);
 
 	/*
 	 * If TIF_SYSCALL_EMU is set, we only get here because of
@@ -1475,6 +1547,6 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs)
 	 * system call instruction.
 	 */
 	if (test_thread_flag(TIF_SINGLESTEP) &&
-	    (current->ptrace & PT_PTRACED))
+	    tracehook_consider_fatal_signal(current, SIGTRAP, SIG_DFL))
 		send_sigtrap(current, regs, 0);
 }
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 724adfc63cb..f4c93f1cfc1 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -29,7 +29,11 @@ EXPORT_SYMBOL(pm_power_off);
 
 static const struct desc_ptr no_idt = {};
 static int reboot_mode;
-enum reboot_type reboot_type = BOOT_KBD;
+/*
+ * Keyboard reset and triple fault may result in INIT, not RESET, which
+ * doesn't work when we're in vmx root mode.  Try ACPI first.
+ */
+enum reboot_type reboot_type = BOOT_ACPI;
 int reboot_force;
 
 #if defined(CONFIG_X86_32) && defined(CONFIG_SMP)
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 673f12cf6eb..46c98efbbf8 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -223,6 +223,9 @@ unsigned long saved_video_mode;
 #define RAMDISK_LOAD_FLAG		0x4000
 
 static char __initdata command_line[COMMAND_LINE_SIZE];
+#ifdef CONFIG_CMDLINE_BOOL
+static char __initdata builtin_cmdline[COMMAND_LINE_SIZE] = CONFIG_CMDLINE;
+#endif
 
 #if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
 struct edd edd;
@@ -665,11 +668,28 @@ void __init setup_arch(char **cmdline_p)
 	bss_resource.start = virt_to_phys(&__bss_start);
 	bss_resource.end = virt_to_phys(&__bss_stop)-1;
 
+#ifdef CONFIG_CMDLINE_BOOL
+#ifdef CONFIG_CMDLINE_OVERRIDE
+	strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE);
+#else
+	if (builtin_cmdline[0]) {
+		/* append boot loader cmdline to builtin */
+		strlcat(builtin_cmdline, " ", COMMAND_LINE_SIZE);
+		strlcat(builtin_cmdline, boot_command_line, COMMAND_LINE_SIZE);
+		strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE);
+	}
+#endif
+#endif
+
 	strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
 	*cmdline_p = command_line;
 
 	parse_early_param();
 
+#ifdef CONFIG_X86_64
+	check_efer();
+#endif
+
 #if defined(CONFIG_VMI) && defined(CONFIG_X86_32)
 	/*
 	 * Must be before kernel pagetables are setup
@@ -738,7 +758,6 @@ void __init setup_arch(char **cmdline_p)
 #else
 	num_physpages = max_pfn;
 
-	check_efer();
  	if (cpu_has_x2apic)
  		check_x2apic();
 
diff --git a/arch/x86/kernel/sigframe.h b/arch/x86/kernel/sigframe.h
index 6dd7e2b70a4..cc673aa55ce 100644
--- a/arch/x86/kernel/sigframe.h
+++ b/arch/x86/kernel/sigframe.h
@@ -34,4 +34,9 @@ struct rt_sigframe {
 	struct siginfo info;
 	/* fp state follows here */
 };
+
+int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+		sigset_t *set, struct pt_regs *regs);
+int ia32_setup_frame(int sig, struct k_sigaction *ka,
+		sigset_t *set, struct pt_regs *regs);
 #endif
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 8d380b699c0..b21070ea33a 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -17,6 +17,7 @@
 #include <linux/errno.h>
 #include <linux/sched.h>
 #include <linux/wait.h>
+#include <linux/tracehook.h>
 #include <linux/elf.h>
 #include <linux/smp.h>
 #include <linux/mm.h>
@@ -556,8 +557,6 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	 * handler too.
 	 */
 	regs->flags &= ~X86_EFLAGS_TF;
-	if (test_thread_flag(TIF_SINGLESTEP))
-		ptrace_notify(SIGTRAP);
 
 	spin_lock_irq(&current->sighand->siglock);
 	sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
@@ -566,6 +565,9 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
+	tracehook_signal_handler(sig, info, ka, regs,
+				 test_thread_flag(TIF_SINGLESTEP));
+
 	return 0;
 }
 
@@ -659,5 +661,10 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 	if (thread_info_flags & _TIF_SIGPENDING)
 		do_signal(regs);
 
+	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
+		clear_thread_flag(TIF_NOTIFY_RESUME);
+		tracehook_notify_resume(regs);
+	}
+
 	clear_thread_flag(TIF_IRET);
 }
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index 4665b598a37..823a55bf8c3 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -15,17 +15,20 @@
 #include <linux/errno.h>
 #include <linux/wait.h>
 #include <linux/ptrace.h>
+#include <linux/tracehook.h>
 #include <linux/unistd.h>
 #include <linux/stddef.h>
 #include <linux/personality.h>
 #include <linux/compiler.h>
+#include <linux/uaccess.h>
+
 #include <asm/processor.h>
 #include <asm/ucontext.h>
-#include <asm/uaccess.h>
 #include <asm/i387.h>
 #include <asm/proto.h>
 #include <asm/ia32_unistd.h>
 #include <asm/mce.h>
+#include <asm/syscall.h>
 #include <asm/syscalls.h>
 #include "sigframe.h"
 
@@ -42,11 +45,6 @@
 # define FIX_EFLAGS	__FIX_EFLAGS
 #endif
 
-int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-               sigset_t *set, struct pt_regs * regs); 
-int ia32_setup_frame(int sig, struct k_sigaction *ka,
-            sigset_t *set, struct pt_regs * regs); 
-
 asmlinkage long
 sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 		struct pt_regs *regs)
@@ -66,7 +64,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
-#define COPY(x)		err |= __get_user(regs->x, &sc->x)
+#define COPY(x)		(err |= __get_user(regs->x, &sc->x))
 
 	COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx);
 	COPY(dx); COPY(cx); COPY(ip);
@@ -96,7 +94,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 	}
 
 	{
-		struct _fpstate __user * buf;
+		struct _fpstate __user *buf;
 		err |= __get_user(buf, &sc->fpstate);
 		err |= restore_i387_xstate(buf);
 	}
@@ -122,7 +120,7 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
 	current->blocked = set;
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
-	
+
 	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax))
 		goto badframe;
 
@@ -132,16 +130,17 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
 	return ax;
 
 badframe:
-	signal_fault(regs,frame,"sigreturn");
+	signal_fault(regs, frame, "sigreturn");
 	return 0;
-}	
+}
 
 /*
  * Set up a signal frame.
  */
 
 static inline int
-setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, unsigned long mask, struct task_struct *me)
+setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
+		unsigned long mask, struct task_struct *me)
 {
 	int err = 0;
 
@@ -197,7 +196,7 @@ get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size)
 }
 
 static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			   sigset_t *set, struct pt_regs * regs)
+			   sigset_t *set, struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
 	void __user *fp = NULL;
@@ -210,19 +209,19 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 			(unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8;
 
 		if (save_i387_xstate(fp) < 0)
-			err |= -1; 
+			err |= -1;
 	} else
 		frame = get_stack(ka, regs, sizeof(struct rt_sigframe)) - 8;
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
 		goto give_sigsegv;
 
-	if (ka->sa.sa_flags & SA_SIGINFO) { 
+	if (ka->sa.sa_flags & SA_SIGINFO) {
 		err |= copy_siginfo_to_user(&frame->info, info);
 		if (err)
 			goto give_sigsegv;
 	}
-		
+
 	/* Create the ucontext.  */
 	if (cpu_has_xsave)
 		err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags);
@@ -235,9 +234,9 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size);
 	err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, set->sig[0], me);
 	err |= __put_user(fp, &frame->uc.uc_mcontext.fpstate);
-	if (sizeof(*set) == 16) { 
+	if (sizeof(*set) == 16) {
 		__put_user(set->sig[0], &frame->uc.uc_sigmask.sig[0]);
-		__put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]); 
+		__put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]);
 	} else
 		err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
 
@@ -248,7 +247,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		err |= __put_user(ka->sa.sa_restorer, &frame->pretcode);
 	} else {
 		/* could use a vstub here */
-		goto give_sigsegv; 
+		goto give_sigsegv;
 	}
 
 	if (err)
@@ -256,7 +255,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 
 	/* Set up registers for signal handler */
 	regs->di = sig;
-	/* In case the signal handler was declared without prototypes */ 
+	/* In case the signal handler was declared without prototypes */
 	regs->ax = 0;
 
 	/* This also works for non SA_SIGINFO handlers because they expect the
@@ -279,37 +278,8 @@ give_sigsegv:
 }
 
 /*
- * Return -1L or the syscall number that @regs is executing.
- */
-static long current_syscall(struct pt_regs *regs)
-{
-	/*
-	 * We always sign-extend a -1 value being set here,
-	 * so this is always either -1L or a syscall number.
-	 */
-	return regs->orig_ax;
-}
-
-/*
- * Return a value that is -EFOO if the system call in @regs->orig_ax
- * returned an error.  This only works for @regs from @current.
- */
-static long current_syscall_ret(struct pt_regs *regs)
-{
-#ifdef CONFIG_IA32_EMULATION
-	if (test_thread_flag(TIF_IA32))
-		/*
-		 * Sign-extend the value so (int)-EFOO becomes (long)-EFOO
-		 * and will match correctly in comparisons.
-		 */
-		return (int) regs->ax;
-#endif
-	return regs->ax;
-}
-
-/*
  * OK, we're invoking a handler
- */	
+ */
 
 static int
 handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
@@ -318,9 +288,9 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	int ret;
 
 	/* Are we from a system call? */
-	if (current_syscall(regs) >= 0) {
+	if (syscall_get_nr(current, regs) >= 0) {
 		/* If so, check system call restarting.. */
-		switch (current_syscall_ret(regs)) {
+		switch (syscall_get_error(current, regs)) {
 		case -ERESTART_RESTARTBLOCK:
 		case -ERESTARTNOHAND:
 			regs->ax = -EINTR;
@@ -353,7 +323,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 			ret = ia32_setup_rt_frame(sig, ka, info, oldset, regs);
 		else
 			ret = ia32_setup_frame(sig, ka, oldset, regs);
-	} else 
+	} else
 #endif
 	ret = setup_rt_frame(sig, ka, info, oldset, regs);
 
@@ -377,15 +347,16 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 		 * handler too.
 		 */
 		regs->flags &= ~X86_EFLAGS_TF;
-		if (test_thread_flag(TIF_SINGLESTEP))
-			ptrace_notify(SIGTRAP);
 
 		spin_lock_irq(&current->sighand->siglock);
-		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
+		sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
 		if (!(ka->sa.sa_flags & SA_NODEFER))
-			sigaddset(&current->blocked,sig);
+			sigaddset(&current->blocked, sig);
 		recalc_sigpending();
 		spin_unlock_irq(&current->sighand->siglock);
+
+		tracehook_signal_handler(sig, info, ka, regs,
+					 test_thread_flag(TIF_SINGLESTEP));
 	}
 
 	return ret;
@@ -442,9 +413,9 @@ static void do_signal(struct pt_regs *regs)
 	}
 
 	/* Did we come from a system call? */
-	if (current_syscall(regs) >= 0) {
+	if (syscall_get_nr(current, regs) >= 0) {
 		/* Restart the system call - no handlers present */
-		switch (current_syscall_ret(regs)) {
+		switch (syscall_get_error(current, regs)) {
 		case -ERESTARTNOHAND:
 		case -ERESTARTSYS:
 		case -ERESTARTNOINTR:
@@ -482,17 +453,23 @@ void do_notify_resume(struct pt_regs *regs, void *unused,
 	/* deal with pending signal delivery */
 	if (thread_info_flags & _TIF_SIGPENDING)
 		do_signal(regs);
+
+	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
+		clear_thread_flag(TIF_NOTIFY_RESUME);
+		tracehook_notify_resume(regs);
+	}
 }
 
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
-{ 
-	struct task_struct *me = current; 
+{
+	struct task_struct *me = current;
 	if (show_unhandled_signals && printk_ratelimit()) {
 		printk("%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx",
-	       me->comm,me->pid,where,frame,regs->ip,regs->sp,regs->orig_ax);
+	       me->comm, me->pid, where, frame, regs->ip,
+		   regs->sp, regs->orig_ax);
 		print_vma_addr(" in ", regs->ip);
 		printk("\n");
 	}
 
-	force_sig(SIGSEGV, me); 
-} 
+	force_sig(SIGSEGV, me);
+}
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index aa804c64b16..9056f7e272c 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -258,6 +258,7 @@ static void __cpuinit smp_callin(void)
 	end_local_APIC_setup();
 	map_cpu_to_logical_apicid();
 
+	notify_cpu_starting(cpuid);
 	/*
 	 * Get our bogomips.
 	 *
@@ -1313,16 +1314,13 @@ __init void prefill_possible_map(void)
 	if (!num_processors)
 		num_processors = 1;
 
-#ifdef CONFIG_HOTPLUG_CPU
 	if (additional_cpus == -1) {
 		if (disabled_cpus > 0)
 			additional_cpus = disabled_cpus;
 		else
 			additional_cpus = 0;
 	}
-#else
-	additional_cpus = 0;
-#endif
+
 	possible = num_processors + additional_cpus;
 	if (possible > NR_CPUS)
 		possible = NR_CPUS;
diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c
index c9288c883e2..6bc211accf0 100644
--- a/arch/x86/kernel/sys_x86_64.c
+++ b/arch/x86/kernel/sys_x86_64.c
@@ -13,16 +13,17 @@
 #include <linux/utsname.h>
 #include <linux/personality.h>
 #include <linux/random.h>
+#include <linux/uaccess.h>
 
-#include <asm/uaccess.h>
 #include <asm/ia32.h>
 #include <asm/syscalls.h>
 
-asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags,
-	unsigned long fd, unsigned long off)
+asmlinkage long sys_mmap(unsigned long addr, unsigned long len,
+		unsigned long prot, unsigned long flags,
+		unsigned long fd, unsigned long off)
 {
 	long error;
-	struct file * file;
+	struct file *file;
 
 	error = -EINVAL;
 	if (off & ~PAGE_MASK)
@@ -57,9 +58,9 @@ static void find_start_end(unsigned long flags, unsigned long *begin,
 		   unmapped base down for this case. This can give
 		   conflicts with the heap, but we assume that glibc
 		   malloc knows how to fall back to mmap. Give it 1GB
-		   of playground for now. -AK */ 
-		*begin = 0x40000000; 
-		*end = 0x80000000;		
+		   of playground for now. -AK */
+		*begin = 0x40000000;
+		*end = 0x80000000;
 		if (current->flags & PF_RANDOMIZE) {
 			new_begin = randomize_range(*begin, *begin + 0x02000000, 0);
 			if (new_begin)
@@ -67,9 +68,9 @@ static void find_start_end(unsigned long flags, unsigned long *begin,
 		}
 	} else {
 		*begin = TASK_UNMAPPED_BASE;
-		*end = TASK_SIZE; 
+		*end = TASK_SIZE;
 	}
-} 
+}
 
 unsigned long
 arch_get_unmapped_area(struct file *filp, unsigned long addr,
@@ -79,11 +80,11 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
 	unsigned long begin, end;
-	
+
 	if (flags & MAP_FIXED)
 		return addr;
 
-	find_start_end(flags, &begin, &end); 
+	find_start_end(flags, &begin, &end);
 
 	if (len > end)
 		return -ENOMEM;
@@ -97,12 +98,12 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	}
 	if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32))
 	    && len <= mm->cached_hole_size) {
-	        mm->cached_hole_size = 0;
+		mm->cached_hole_size = 0;
 		mm->free_area_cache = begin;
 	}
 	addr = mm->free_area_cache;
-	if (addr < begin) 
-		addr = begin; 
+	if (addr < begin)
+		addr = begin;
 	start_addr = addr;
 
 full_search:
@@ -128,7 +129,7 @@ full_search:
 			return addr;
 		}
 		if (addr + mm->cached_hole_size < vma->vm_start)
-		        mm->cached_hole_size = vma->vm_start - addr;
+			mm->cached_hole_size = vma->vm_start - addr;
 
 		addr = vma->vm_end;
 	}
@@ -178,7 +179,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 		vma = find_vma(mm, addr-len);
 		if (!vma || addr <= vma->vm_start)
 			/* remember the address as a hint for next time */
-			return (mm->free_area_cache = addr-len);
+			return mm->free_area_cache = addr-len;
 	}
 
 	if (mm->mmap_base < len)
@@ -195,7 +196,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 		vma = find_vma(mm, addr);
 		if (!vma || addr+len <= vma->vm_start)
 			/* remember the address as a hint for next time */
-			return (mm->free_area_cache = addr);
+			return mm->free_area_cache = addr;
 
 		/* remember the largest hole we saw so far */
 		if (addr + mm->cached_hole_size < vma->vm_start)
@@ -225,13 +226,13 @@ bottomup:
 }
 
 
-asmlinkage long sys_uname(struct new_utsname __user * name)
+asmlinkage long sys_uname(struct new_utsname __user *name)
 {
 	int err;
 	down_read(&uts_sem);
-	err = copy_to_user(name, utsname(), sizeof (*name));
+	err = copy_to_user(name, utsname(), sizeof(*name));
 	up_read(&uts_sem);
-	if (personality(current->personality) == PER_LINUX32) 
-		err |= copy_to_user(&name->machine, "i686", 5); 		
+	if (personality(current->personality) == PER_LINUX32)
+		err |= copy_to_user(&name->machine, "i686", 5);
 	return err ? -EFAULT : 0;
 }
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index b42068fb7b7..2887a789e38 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -32,6 +32,8 @@
 #include <linux/bug.h>
 #include <linux/nmi.h>
 #include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/io.h>
 
 #if defined(CONFIG_EDAC)
 #include <linux/edac.h>
@@ -45,9 +47,6 @@
 #include <asm/unwind.h>
 #include <asm/desc.h>
 #include <asm/i387.h>
-#include <asm/nmi.h>
-#include <asm/smp.h>
-#include <asm/io.h>
 #include <asm/pgalloc.h>
 #include <asm/proto.h>
 #include <asm/pda.h>
@@ -85,7 +84,8 @@ static inline void preempt_conditional_cli(struct pt_regs *regs)
 
 void printk_address(unsigned long address, int reliable)
 {
-	printk(" [<%016lx>] %s%pS\n", address, reliable ? "": "? ", (void *) address);
+	printk(" [<%016lx>] %s%pS\n",
+			address, reliable ?	"" : "? ", (void *) address);
 }
 
 static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
@@ -98,7 +98,8 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 		[STACKFAULT_STACK - 1] = "#SS",
 		[MCE_STACK - 1] = "#MC",
 #if DEBUG_STKSZ > EXCEPTION_STKSZ
-		[N_EXCEPTION_STACKS ... N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]"
+		[N_EXCEPTION_STACKS ...
+			N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]"
 #endif
 	};
 	unsigned k;
@@ -163,7 +164,7 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 }
 
 /*
- * x86-64 can have up to three kernel stacks: 
+ * x86-64 can have up to three kernel stacks:
  * process stack
  * interrupt stack
  * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
@@ -219,7 +220,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 		const struct stacktrace_ops *ops, void *data)
 {
 	const unsigned cpu = get_cpu();
-	unsigned long *irqstack_end = (unsigned long*)cpu_pda(cpu)->irqstackptr;
+	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
 	unsigned used = 0;
 	struct thread_info *tinfo;
 
@@ -237,7 +238,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 	if (!bp) {
 		if (task == current) {
 			/* Grab bp right from our regs */
-			asm("movq %%rbp, %0" : "=r" (bp) :);
+			asm("movq %%rbp, %0" : "=r" (bp) : );
 		} else {
 			/* bp is the last reg pushed by switch_to */
 			bp = *(unsigned long *) task->thread.sp;
@@ -356,11 +357,15 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
 	unsigned long *stack;
 	int i;
 	const int cpu = smp_processor_id();
-	unsigned long *irqstack_end = (unsigned long *) (cpu_pda(cpu)->irqstackptr);
-	unsigned long *irqstack = (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
+	unsigned long *irqstack_end =
+		(unsigned long *) (cpu_pda(cpu)->irqstackptr);
+	unsigned long *irqstack =
+		(unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
 
-	// debugging aid: "show_stack(NULL, NULL);" prints the
-	// back trace for this cpu.
+	/*
+	 * debugging aid: "show_stack(NULL, NULL);" prints the
+	 * back trace for this cpu.
+	 */
 
 	if (sp == NULL) {
 		if (task)
@@ -404,7 +409,7 @@ void dump_stack(void)
 
 #ifdef CONFIG_FRAME_POINTER
 	if (!bp)
-		asm("movq %%rbp, %0" : "=r" (bp):);
+		asm("movq %%rbp, %0" : "=r" (bp) : );
 #endif
 
 	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
@@ -414,7 +419,6 @@ void dump_stack(void)
 		init_utsname()->version);
 	show_trace(NULL, NULL, &stack, bp);
 }
-
 EXPORT_SYMBOL(dump_stack);
 
 void show_registers(struct pt_regs *regs)
@@ -492,7 +496,7 @@ unsigned __kprobes long oops_begin(void)
 	raw_local_irq_save(flags);
 	cpu = smp_processor_id();
 	if (!__raw_spin_trylock(&die_lock)) {
-		if (cpu == die_owner) 
+		if (cpu == die_owner)
 			/* nested oops. should stop eventually */;
 		else
 			__raw_spin_lock(&die_lock);
@@ -637,7 +641,7 @@ kernel_trap:
 }
 
 #define DO_ERROR(trapnr, signr, str, name) \
-asmlinkage void do_##name(struct pt_regs * regs, long error_code)	\
+asmlinkage void do_##name(struct pt_regs *regs, long error_code)	\
 {									\
 	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)	\
 							== NOTIFY_STOP)	\
@@ -647,7 +651,7 @@ asmlinkage void do_##name(struct pt_regs * regs, long error_code)	\
 }
 
 #define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr)		\
-asmlinkage void do_##name(struct pt_regs * regs, long error_code)	\
+asmlinkage void do_##name(struct pt_regs *regs, long error_code)	\
 {									\
 	siginfo_t info;							\
 	info.si_signo = signr;						\
@@ -682,7 +686,7 @@ asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code)
 	preempt_conditional_cli(regs);
 }
 
-asmlinkage void do_double_fault(struct pt_regs * regs, long error_code)
+asmlinkage void do_double_fault(struct pt_regs *regs, long error_code)
 {
 	static const char str[] = "double fault";
 	struct task_struct *tsk = current;
@@ -777,9 +781,10 @@ io_check_error(unsigned char reason, struct pt_regs *regs)
 }
 
 static notrace __kprobes void
-unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
+unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 {
-	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
+	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) ==
+			NOTIFY_STOP)
 		return;
 	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
 		reason);
@@ -881,7 +886,7 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 	else if (user_mode(eregs))
 		regs = task_pt_regs(current);
 	/* Exception from kernel and interrupts are enabled. Move to
- 	   kernel process stack. */
+	   kernel process stack. */
 	else if (eregs->flags & X86_EFLAGS_IF)
 		regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs));
 	if (eregs != regs)
@@ -890,7 +895,7 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 }
 
 /* runs on IST stack. */
-asmlinkage void __kprobes do_debug(struct pt_regs * regs,
+asmlinkage void __kprobes do_debug(struct pt_regs *regs,
 				   unsigned long error_code)
 {
 	struct task_struct *tsk = current;
@@ -1034,7 +1039,7 @@ asmlinkage void do_coprocessor_error(struct pt_regs *regs)
 
 asmlinkage void bad_intr(void)
 {
-	printk("bad interrupt"); 
+	printk("bad interrupt");
 }
 
 asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
@@ -1046,7 +1051,7 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
 
 	conditional_sti(regs);
 	if (!user_mode(regs) &&
-        	kernel_math_error(regs, "kernel simd math error", 19))
+			kernel_math_error(regs, "kernel simd math error", 19))
 		return;
 
 	/*
@@ -1091,7 +1096,7 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
 	force_sig_info(SIGFPE, &info, task);
 }
 
-asmlinkage void do_spurious_interrupt_bug(struct pt_regs * regs)
+asmlinkage void do_spurious_interrupt_bug(struct pt_regs *regs)
 {
 }
 
@@ -1148,8 +1153,10 @@ void __init trap_init(void)
 	set_intr_gate(0, &divide_error);
 	set_intr_gate_ist(1, &debug, DEBUG_STACK);
 	set_intr_gate_ist(2, &nmi, NMI_STACK);
- 	set_system_gate_ist(3, &int3, DEBUG_STACK); /* int3 can be called from all */
-	set_system_gate(4, &overflow); /* int4 can be called from all */
+	/* int3 can be called from all */
+	set_system_gate_ist(3, &int3, DEBUG_STACK);
+	/* int4 can be called from all */
+	set_system_gate(4, &overflow);
 	set_intr_gate(5, &bounds);
 	set_intr_gate(6, &invalid_op);
 	set_intr_gate(7, &device_not_available);
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 346cae5ac42..161bb850fc4 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -104,7 +104,7 @@ __setup("notsc", notsc_setup);
 /*
  * Read TSC and the reference counters. Take care of SMI disturbance
  */
-static u64 tsc_read_refs(u64 *pm, u64 *hpet)
+static u64 tsc_read_refs(u64 *p, int hpet)
 {
 	u64 t1, t2;
 	int i;
@@ -112,9 +112,9 @@ static u64 tsc_read_refs(u64 *pm, u64 *hpet)
 	for (i = 0; i < MAX_RETRIES; i++) {
 		t1 = get_cycles();
 		if (hpet)
-			*hpet = hpet_readl(HPET_COUNTER) & 0xFFFFFFFF;
+			*p = hpet_readl(HPET_COUNTER) & 0xFFFFFFFF;
 		else
-			*pm = acpi_pm_read_early();
+			*p = acpi_pm_read_early();
 		t2 = get_cycles();
 		if ((t2 - t1) < SMI_TRESHOLD)
 			return t2;
@@ -123,13 +123,59 @@ static u64 tsc_read_refs(u64 *pm, u64 *hpet)
 }
 
 /*
+ * Calculate the TSC frequency from HPET reference
+ */
+static unsigned long calc_hpet_ref(u64 deltatsc, u64 hpet1, u64 hpet2)
+{
+	u64 tmp;
+
+	if (hpet2 < hpet1)
+		hpet2 += 0x100000000ULL;
+	hpet2 -= hpet1;
+	tmp = ((u64)hpet2 * hpet_readl(HPET_PERIOD));
+	do_div(tmp, 1000000);
+	do_div(deltatsc, tmp);
+
+	return (unsigned long) deltatsc;
+}
+
+/*
+ * Calculate the TSC frequency from PMTimer reference
+ */
+static unsigned long calc_pmtimer_ref(u64 deltatsc, u64 pm1, u64 pm2)
+{
+	u64 tmp;
+
+	if (!pm1 && !pm2)
+		return ULONG_MAX;
+
+	if (pm2 < pm1)
+		pm2 += (u64)ACPI_PM_OVRRUN;
+	pm2 -= pm1;
+	tmp = pm2 * 1000000000LL;
+	do_div(tmp, PMTMR_TICKS_PER_SEC);
+	do_div(deltatsc, tmp);
+
+	return (unsigned long) deltatsc;
+}
+
+#define CAL_MS		10
+#define CAL_LATCH	(CLOCK_TICK_RATE / (1000 / CAL_MS))
+#define CAL_PIT_LOOPS	1000
+
+#define CAL2_MS		50
+#define CAL2_LATCH	(CLOCK_TICK_RATE / (1000 / CAL2_MS))
+#define CAL2_PIT_LOOPS	5000
+
+
+/*
  * Try to calibrate the TSC against the Programmable
  * Interrupt Timer and return the frequency of the TSC
  * in kHz.
  *
  * Return ULONG_MAX on failure to calibrate.
  */
-static unsigned long pit_calibrate_tsc(void)
+static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin)
 {
 	u64 tsc, t1, t2, delta;
 	unsigned long tscmin, tscmax;
@@ -144,8 +190,8 @@ static unsigned long pit_calibrate_tsc(void)
 	 * (LSB then MSB) to begin countdown.
 	 */
 	outb(0xb0, 0x43);
-	outb((CLOCK_TICK_RATE / (1000 / 50)) & 0xff, 0x42);
-	outb((CLOCK_TICK_RATE / (1000 / 50)) >> 8, 0x42);
+	outb(latch & 0xff, 0x42);
+	outb(latch >> 8, 0x42);
 
 	tsc = t1 = t2 = get_cycles();
 
@@ -166,31 +212,154 @@ static unsigned long pit_calibrate_tsc(void)
 	/*
 	 * Sanity checks:
 	 *
-	 * If we were not able to read the PIT more than 5000
+	 * If we were not able to read the PIT more than loopmin
 	 * times, then we have been hit by a massive SMI
 	 *
 	 * If the maximum is 10 times larger than the minimum,
 	 * then we got hit by an SMI as well.
 	 */
-	if (pitcnt < 5000 || tscmax > 10 * tscmin)
+	if (pitcnt < loopmin || tscmax > 10 * tscmin)
 		return ULONG_MAX;
 
 	/* Calculate the PIT value */
 	delta = t2 - t1;
-	do_div(delta, 50);
+	do_div(delta, ms);
 	return delta;
 }
 
+/*
+ * This reads the current MSB of the PIT counter, and
+ * checks if we are running on sufficiently fast and
+ * non-virtualized hardware.
+ *
+ * Our expectations are:
+ *
+ *  - the PIT is running at roughly 1.19MHz
+ *
+ *  - each IO is going to take about 1us on real hardware,
+ *    but we allow it to be much faster (by a factor of 10) or
+ *    _slightly_ slower (ie we allow up to a 2us read+counter
+ *    update - anything else implies a unacceptably slow CPU
+ *    or PIT for the fast calibration to work.
+ *
+ *  - with 256 PIT ticks to read the value, we have 214us to
+ *    see the same MSB (and overhead like doing a single TSC
+ *    read per MSB value etc).
+ *
+ *  - We're doing 2 reads per loop (LSB, MSB), and we expect
+ *    them each to take about a microsecond on real hardware.
+ *    So we expect a count value of around 100. But we'll be
+ *    generous, and accept anything over 50.
+ *
+ *  - if the PIT is stuck, and we see *many* more reads, we
+ *    return early (and the next caller of pit_expect_msb()
+ *    then consider it a failure when they don't see the
+ *    next expected value).
+ *
+ * These expectations mean that we know that we have seen the
+ * transition from one expected value to another with a fairly
+ * high accuracy, and we didn't miss any events. We can thus
+ * use the TSC value at the transitions to calculate a pretty
+ * good value for the TSC frequencty.
+ */
+static inline int pit_expect_msb(unsigned char val)
+{
+	int count = 0;
+
+	for (count = 0; count < 50000; count++) {
+		/* Ignore LSB */
+		inb(0x42);
+		if (inb(0x42) != val)
+			break;
+	}
+	return count > 50;
+}
+
+/*
+ * How many MSB values do we want to see? We aim for a
+ * 15ms calibration, which assuming a 2us counter read
+ * error should give us roughly 150 ppm precision for
+ * the calibration.
+ */
+#define QUICK_PIT_MS 15
+#define QUICK_PIT_ITERATIONS (QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256)
+
+static unsigned long quick_pit_calibrate(void)
+{
+	/* Set the Gate high, disable speaker */
+	outb((inb(0x61) & ~0x02) | 0x01, 0x61);
+
+	/*
+	 * Counter 2, mode 0 (one-shot), binary count
+	 *
+	 * NOTE! Mode 2 decrements by two (and then the
+	 * output is flipped each time, giving the same
+	 * final output frequency as a decrement-by-one),
+	 * so mode 0 is much better when looking at the
+	 * individual counts.
+	 */
+	outb(0xb0, 0x43);
+
+	/* Start at 0xffff */
+	outb(0xff, 0x42);
+	outb(0xff, 0x42);
+
+	if (pit_expect_msb(0xff)) {
+		int i;
+		u64 t1, t2, delta;
+		unsigned char expect = 0xfe;
+
+		t1 = get_cycles();
+		for (i = 0; i < QUICK_PIT_ITERATIONS; i++, expect--) {
+			if (!pit_expect_msb(expect))
+				goto failed;
+		}
+		t2 = get_cycles();
+
+		/*
+		 * Make sure we can rely on the second TSC timestamp:
+		 */
+		if (!pit_expect_msb(expect))
+			goto failed;
+
+		/*
+		 * Ok, if we get here, then we've seen the
+		 * MSB of the PIT decrement QUICK_PIT_ITERATIONS
+		 * times, and each MSB had many hits, so we never
+		 * had any sudden jumps.
+		 *
+		 * As a result, we can depend on there not being
+		 * any odd delays anywhere, and the TSC reads are
+		 * reliable.
+		 *
+		 * kHz = ticks / time-in-seconds / 1000;
+		 * kHz = (t2 - t1) / (QPI * 256 / PIT_TICK_RATE) / 1000
+		 * kHz = ((t2 - t1) * PIT_TICK_RATE) / (QPI * 256 * 1000)
+		 */
+		delta = (t2 - t1)*PIT_TICK_RATE;
+		do_div(delta, QUICK_PIT_ITERATIONS*256*1000);
+		printk("Fast TSC calibration using PIT\n");
+		return delta;
+	}
+failed:
+	return 0;
+}
 
 /**
  * native_calibrate_tsc - calibrate the tsc on boot
  */
 unsigned long native_calibrate_tsc(void)
 {
-	u64 tsc1, tsc2, delta, pm1, pm2, hpet1, hpet2;
+	u64 tsc1, tsc2, delta, ref1, ref2;
 	unsigned long tsc_pit_min = ULONG_MAX, tsc_ref_min = ULONG_MAX;
-	unsigned long flags;
-	int hpet = is_hpet_enabled(), i;
+	unsigned long flags, latch, ms, fast_calibrate;
+	int hpet = is_hpet_enabled(), i, loopmin;
+
+	local_irq_save(flags);
+	fast_calibrate = quick_pit_calibrate();
+	local_irq_restore(flags);
+	if (fast_calibrate)
+		return fast_calibrate;
 
 	/*
 	 * Run 5 calibration loops to get the lowest frequency value
@@ -216,7 +385,13 @@ unsigned long native_calibrate_tsc(void)
 	 * calibration delay loop as we have to wait for a certain
 	 * amount of time anyway.
 	 */
-	for (i = 0; i < 5; i++) {
+
+	/* Preset PIT loop values */
+	latch = CAL_LATCH;
+	ms = CAL_MS;
+	loopmin = CAL_PIT_LOOPS;
+
+	for (i = 0; i < 3; i++) {
 		unsigned long tsc_pit_khz;
 
 		/*
@@ -226,16 +401,16 @@ unsigned long native_calibrate_tsc(void)
 		 * read the end value.
 		 */
 		local_irq_save(flags);
-		tsc1 = tsc_read_refs(&pm1, hpet ? &hpet1 : NULL);
-		tsc_pit_khz = pit_calibrate_tsc();
-		tsc2 = tsc_read_refs(&pm2, hpet ? &hpet2 : NULL);
+		tsc1 = tsc_read_refs(&ref1, hpet);
+		tsc_pit_khz = pit_calibrate_tsc(latch, ms, loopmin);
+		tsc2 = tsc_read_refs(&ref2, hpet);
 		local_irq_restore(flags);
 
 		/* Pick the lowest PIT TSC calibration so far */
 		tsc_pit_min = min(tsc_pit_min, tsc_pit_khz);
 
 		/* hpet or pmtimer available ? */
-		if (!hpet && !pm1 && !pm2)
+		if (!hpet && !ref1 && !ref2)
 			continue;
 
 		/* Check, whether the sampling was disturbed by an SMI */
@@ -243,23 +418,41 @@ unsigned long native_calibrate_tsc(void)
 			continue;
 
 		tsc2 = (tsc2 - tsc1) * 1000000LL;
+		if (hpet)
+			tsc2 = calc_hpet_ref(tsc2, ref1, ref2);
+		else
+			tsc2 = calc_pmtimer_ref(tsc2, ref1, ref2);
 
-		if (hpet) {
-			if (hpet2 < hpet1)
-				hpet2 += 0x100000000ULL;
-			hpet2 -= hpet1;
-			tsc1 = ((u64)hpet2 * hpet_readl(HPET_PERIOD));
-			do_div(tsc1, 1000000);
-		} else {
-			if (pm2 < pm1)
-				pm2 += (u64)ACPI_PM_OVRRUN;
-			pm2 -= pm1;
-			tsc1 = pm2 * 1000000000LL;
-			do_div(tsc1, PMTMR_TICKS_PER_SEC);
+		tsc_ref_min = min(tsc_ref_min, (unsigned long) tsc2);
+
+		/* Check the reference deviation */
+		delta = ((u64) tsc_pit_min) * 100;
+		do_div(delta, tsc_ref_min);
+
+		/*
+		 * If both calibration results are inside a 10% window
+		 * then we can be sure, that the calibration
+		 * succeeded. We break out of the loop right away. We
+		 * use the reference value, as it is more precise.
+		 */
+		if (delta >= 90 && delta <= 110) {
+			printk(KERN_INFO
+			       "TSC: PIT calibration matches %s. %d loops\n",
+			       hpet ? "HPET" : "PMTIMER", i + 1);
+			return tsc_ref_min;
 		}
 
-		do_div(tsc2, tsc1);
-		tsc_ref_min = min(tsc_ref_min, (unsigned long) tsc2);
+		/*
+		 * Check whether PIT failed more than once. This
+		 * happens in virtualized environments. We need to
+		 * give the virtual PC a slightly longer timeframe for
+		 * the HPET/PMTIMER to make the result precise.
+		 */
+		if (i == 1 && tsc_pit_min == ULONG_MAX) {
+			latch = CAL2_LATCH;
+			ms = CAL2_MS;
+			loopmin = CAL2_PIT_LOOPS;
+		}
 	}
 
 	/*
@@ -267,11 +460,10 @@ unsigned long native_calibrate_tsc(void)
 	 */
 	if (tsc_pit_min == ULONG_MAX) {
 		/* PIT gave no useful value */
-		printk(KERN_WARNING "TSC: PIT calibration failed due to "
-		       "SMI disturbance.\n");
+		printk(KERN_WARNING "TSC: Unable to calibrate against PIT\n");
 
 		/* We don't have an alternative source, disable TSC */
-		if (!hpet && !pm1 && !pm2) {
+		if (!hpet && !ref1 && !ref2) {
 			printk("TSC: No reference (HPET/PMTIMER) available\n");
 			return 0;
 		}
@@ -279,7 +471,7 @@ unsigned long native_calibrate_tsc(void)
 		/* The alternative source failed as well, disable TSC */
 		if (tsc_ref_min == ULONG_MAX) {
 			printk(KERN_WARNING "TSC: HPET/PMTIMER calibration "
-			       "failed due to SMI disturbance.\n");
+			       "failed.\n");
 			return 0;
 		}
 
@@ -291,44 +483,25 @@ unsigned long native_calibrate_tsc(void)
 	}
 
 	/* We don't have an alternative source, use the PIT calibration value */
-	if (!hpet && !pm1 && !pm2) {
+	if (!hpet && !ref1 && !ref2) {
 		printk(KERN_INFO "TSC: Using PIT calibration value\n");
 		return tsc_pit_min;
 	}
 
 	/* The alternative source failed, use the PIT calibration value */
 	if (tsc_ref_min == ULONG_MAX) {
-		printk(KERN_WARNING "TSC: HPET/PMTIMER calibration failed due "
-		       "to SMI disturbance. Using PIT calibration\n");
+		printk(KERN_WARNING "TSC: HPET/PMTIMER calibration failed. "
+		       "Using PIT calibration\n");
 		return tsc_pit_min;
 	}
 
-	/* Check the reference deviation */
-	delta = ((u64) tsc_pit_min) * 100;
-	do_div(delta, tsc_ref_min);
-
-	/*
-	 * If both calibration results are inside a 5% window, the we
-	 * use the lower frequency of those as it is probably the
-	 * closest estimate.
-	 */
-	if (delta >= 95 && delta <= 105) {
-		printk(KERN_INFO "TSC: PIT calibration confirmed by %s.\n",
-		       hpet ? "HPET" : "PMTIMER");
-		printk(KERN_INFO "TSC: using %s calibration value\n",
-		       tsc_pit_min <= tsc_ref_min ? "PIT" :
-		       hpet ? "HPET" : "PMTIMER");
-		return tsc_pit_min <= tsc_ref_min ? tsc_pit_min : tsc_ref_min;
-	}
-
-	printk(KERN_WARNING "TSC: PIT calibration deviates from %s: %lu %lu.\n",
-	       hpet ? "HPET" : "PMTIMER", tsc_pit_min, tsc_ref_min);
-
 	/*
 	 * The calibration values differ too much. In doubt, we use
 	 * the PIT value as we know that there are PMTIMERs around
-	 * running at double speed.
+	 * running at double speed. At least we let the user know:
 	 */
+	printk(KERN_WARNING "TSC: PIT calibration deviates from %s: %lu %lu.\n",
+	       hpet ? "HPET" : "PMTIMER", tsc_pit_min, tsc_ref_min);
 	printk(KERN_INFO "TSC: Using PIT calibration value\n");
 	return tsc_pit_min;
 }
diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c
index 594ef47f0a6..61a97e616f7 100644
--- a/arch/x86/kernel/visws_quirks.c
+++ b/arch/x86/kernel/visws_quirks.c
@@ -25,45 +25,31 @@
 #include <asm/visws/cobalt.h>
 #include <asm/visws/piix4.h>
 #include <asm/arch_hooks.h>
+#include <asm/io_apic.h>
 #include <asm/fixmap.h>
 #include <asm/reboot.h>
 #include <asm/setup.h>
 #include <asm/e820.h>
-#include <asm/smp.h>
 #include <asm/io.h>
 
 #include <mach_ipi.h>
 
 #include "mach_apic.h"
 
-#include <linux/init.h>
-#include <linux/smp.h>
-
 #include <linux/kernel_stat.h>
-#include <linux/interrupt.h>
-#include <linux/init.h>
 
-#include <asm/io.h>
-#include <asm/apic.h>
 #include <asm/i8259.h>
 #include <asm/irq_vectors.h>
-#include <asm/visws/cobalt.h>
 #include <asm/visws/lithium.h>
-#include <asm/visws/piix4.h>
 
 #include <linux/sched.h>
 #include <linux/kernel.h>
-#include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/pci_ids.h>
 
 extern int no_broadcast;
 
-#include <asm/io.h>
 #include <asm/apic.h>
-#include <asm/arch_hooks.h>
-#include <asm/visws/cobalt.h>
-#include <asm/visws/lithium.h>
 
 char visws_board_type	= -1;
 char visws_board_rev	= -1;
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 61531d5c950..8b6c393ab9f 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -235,7 +235,7 @@ static void vmi_write_ldt_entry(struct desc_struct *dt, int entry,
 				const void *desc)
 {
 	u32 *ldt_entry = (u32 *)desc;
-	vmi_ops.write_idt_entry(dt, entry, ldt_entry[0], ldt_entry[1]);
+	vmi_ops.write_ldt_entry(dt, entry, ldt_entry[0], ldt_entry[1]);
 }
 
 static void vmi_load_sp0(struct tss_struct *tss,
@@ -393,13 +393,13 @@ static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type)
 }
 #endif
 
-static void vmi_allocate_pte(struct mm_struct *mm, u32 pfn)
+static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn)
 {
 	vmi_set_page_type(pfn, VMI_PAGE_L1);
 	vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0);
 }
 
-static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn)
+static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn)
 {
  	/*
 	 * This call comes in very early, before mem_map is setup.
@@ -410,20 +410,20 @@ static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn)
 	vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0);
 }
 
-static void vmi_allocate_pmd_clone(u32 pfn, u32 clonepfn, u32 start, u32 count)
+static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count)
 {
  	vmi_set_page_type(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE);
 	vmi_check_page_type(clonepfn, VMI_PAGE_L2);
 	vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count);
 }
 
-static void vmi_release_pte(u32 pfn)
+static void vmi_release_pte(unsigned long pfn)
 {
 	vmi_ops.release_page(pfn, VMI_PAGE_L1);
 	vmi_set_page_type(pfn, VMI_PAGE_NORMAL);
 }
 
-static void vmi_release_pmd(u32 pfn)
+static void vmi_release_pmd(unsigned long pfn)
 {
 	vmi_ops.release_page(pfn, VMI_PAGE_L2);
 	vmi_set_page_type(pfn, VMI_PAGE_NORMAL);
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index 0c029e8959c..7766d36983f 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -61,7 +61,7 @@ static void vsmp_irq_enable(void)
 	native_restore_fl((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC));
 }
 
-static unsigned __init vsmp_patch(u8 type, u16 clobbers, void *ibuf,
+static unsigned __init_or_module vsmp_patch(u8 type, u16 clobbers, void *ibuf,
 				  unsigned long addr, unsigned len)
 {
 	switch (type) {