Merge commit 'origin/master' into next

55 files changed, 1525 insertions, 666 deletions

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 65d0b72777ea..29dc0c89d4af 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -538,9 +538,10 @@ static int __cpuinit _acpi_map_lsapic(acpi_handle handle, int *pcpu)
 	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
 	union acpi_object *obj;
 	struct acpi_madt_local_apic *lapic;
-	cpumask_t tmp_map, new_map;
+	cpumask_var_t tmp_map, new_map;
 	u8 physid;
 	int cpu;
+	int retval = -ENOMEM;
 
 	if (ACPI_FAILURE(acpi_evaluate_object(handle, "_MAT", NULL, &buffer)))
 		return -EINVAL;
@@ -569,23 +570,37 @@ static int __cpuinit _acpi_map_lsapic(acpi_handle handle, int *pcpu)
 	buffer.length = ACPI_ALLOCATE_BUFFER;
 	buffer.pointer = NULL;
 
-	tmp_map = cpu_present_map;
+	if (!alloc_cpumask_var(&tmp_map, GFP_KERNEL))
+		goto out;
+
+	if (!alloc_cpumask_var(&new_map, GFP_KERNEL))
+		goto free_tmp_map;
+
+	cpumask_copy(tmp_map, cpu_present_mask);
 	acpi_register_lapic(physid, lapic->lapic_flags & ACPI_MADT_ENABLED);
 
 	/*
 	 * If mp_register_lapic successfully generates a new logical cpu
 	 * number, then the following will get us exactly what was mapped
 	 */
-	cpus_andnot(new_map, cpu_present_map, tmp_map);
-	if (cpus_empty(new_map)) {
+	cpumask_andnot(new_map, cpu_present_mask, tmp_map);
+	if (cpumask_empty(new_map)) {
 		printk ("Unable to map lapic to logical cpu number\n");
-		return -EINVAL;
+		retval = -EINVAL;
+		goto free_new_map;
 	}
 
-	cpu = first_cpu(new_map);
+	cpu = cpumask_first(new_map);
 
 	*pcpu = cpu;
-	return 0;
+	retval = 0;
+
+free_new_map:
+	free_cpumask_var(new_map);
+free_tmp_map:
+	free_cpumask_var(tmp_map);
+out:
+	return retval;
 }
 
 /* wrapper to silence section mismatch warning */
@@ -598,7 +613,7 @@ EXPORT_SYMBOL(acpi_map_lsapic);
 int acpi_unmap_lsapic(int cpu)
 {
 	per_cpu(x86_cpu_to_apicid, cpu) = -1;
-	cpu_clear(cpu, cpu_present_map);
+	set_cpu_present(cpu, false);
 	num_processors--;
 
 	return (0);
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 2e2da717b350..5113c080f0c4 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -20,8 +20,12 @@
 #include <linux/pci.h>
 #include <linux/gfp.h>
 #include <linux/bitops.h>
+#include <linux/debugfs.h>
 #include <linux/scatterlist.h>
 #include <linux/iommu-helper.h>
+#ifdef CONFIG_IOMMU_API
+#include <linux/iommu.h>
+#endif
 #include <asm/proto.h>
 #include <asm/iommu.h>
 #include <asm/gart.h>
@@ -38,6 +42,10 @@ static DEFINE_RWLOCK(amd_iommu_devtable_lock);
 static LIST_HEAD(iommu_pd_list);
 static DEFINE_SPINLOCK(iommu_pd_list_lock);
 
+#ifdef CONFIG_IOMMU_API
+static struct iommu_ops amd_iommu_ops;
+#endif
+
 /*
  * general struct to manage commands send to an IOMMU
  */
@@ -47,6 +55,68 @@ struct iommu_cmd {
 
 static int dma_ops_unity_map(struct dma_ops_domain *dma_dom,
 			     struct unity_map_entry *e);
+static struct dma_ops_domain *find_protection_domain(u16 devid);
+
+
+#ifdef CONFIG_AMD_IOMMU_STATS
+
+/*
+ * Initialization code for statistics collection
+ */
+
+DECLARE_STATS_COUNTER(compl_wait);
+DECLARE_STATS_COUNTER(cnt_map_single);
+DECLARE_STATS_COUNTER(cnt_unmap_single);
+DECLARE_STATS_COUNTER(cnt_map_sg);
+DECLARE_STATS_COUNTER(cnt_unmap_sg);
+DECLARE_STATS_COUNTER(cnt_alloc_coherent);
+DECLARE_STATS_COUNTER(cnt_free_coherent);
+DECLARE_STATS_COUNTER(cross_page);
+DECLARE_STATS_COUNTER(domain_flush_single);
+DECLARE_STATS_COUNTER(domain_flush_all);
+DECLARE_STATS_COUNTER(alloced_io_mem);
+DECLARE_STATS_COUNTER(total_map_requests);
+
+static struct dentry *stats_dir;
+static struct dentry *de_isolate;
+static struct dentry *de_fflush;
+
+static void amd_iommu_stats_add(struct __iommu_counter *cnt)
+{
+	if (stats_dir == NULL)
+		return;
+
+	cnt->dent = debugfs_create_u64(cnt->name, 0444, stats_dir,
+				       &cnt->value);
+}
+
+static void amd_iommu_stats_init(void)
+{
+	stats_dir = debugfs_create_dir("amd-iommu", NULL);
+	if (stats_dir == NULL)
+		return;
+
+	de_isolate = debugfs_create_bool("isolation", 0444, stats_dir,
+					 (u32 *)&amd_iommu_isolate);
+
+	de_fflush  = debugfs_create_bool("fullflush", 0444, stats_dir,
+					 (u32 *)&amd_iommu_unmap_flush);
+
+	amd_iommu_stats_add(&compl_wait);
+	amd_iommu_stats_add(&cnt_map_single);
+	amd_iommu_stats_add(&cnt_unmap_single);
+	amd_iommu_stats_add(&cnt_map_sg);
+	amd_iommu_stats_add(&cnt_unmap_sg);
+	amd_iommu_stats_add(&cnt_alloc_coherent);
+	amd_iommu_stats_add(&cnt_free_coherent);
+	amd_iommu_stats_add(&cross_page);
+	amd_iommu_stats_add(&domain_flush_single);
+	amd_iommu_stats_add(&domain_flush_all);
+	amd_iommu_stats_add(&alloced_io_mem);
+	amd_iommu_stats_add(&total_map_requests);
+}
+
+#endif
 
 /* returns !0 if the IOMMU is caching non-present entries in its TLB */
 static int iommu_has_npcache(struct amd_iommu *iommu)
@@ -189,13 +259,55 @@ static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd)
 	spin_lock_irqsave(&iommu->lock, flags);
 	ret = __iommu_queue_command(iommu, cmd);
 	if (!ret)
-		iommu->need_sync = 1;
+		iommu->need_sync = true;
 	spin_unlock_irqrestore(&iommu->lock, flags);
 
 	return ret;
 }
 
 /*
+ * This function waits until an IOMMU has completed a completion
+ * wait command
+ */
+static void __iommu_wait_for_completion(struct amd_iommu *iommu)
+{
+	int ready = 0;
+	unsigned status = 0;
+	unsigned long i = 0;
+
+	INC_STATS_COUNTER(compl_wait);
+
+	while (!ready && (i < EXIT_LOOP_COUNT)) {
+		++i;
+		/* wait for the bit to become one */
+		status = readl(iommu->mmio_base + MMIO_STATUS_OFFSET);
+		ready = status & MMIO_STATUS_COM_WAIT_INT_MASK;
+	}
+
+	/* set bit back to zero */
+	status &= ~MMIO_STATUS_COM_WAIT_INT_MASK;
+	writel(status, iommu->mmio_base + MMIO_STATUS_OFFSET);
+
+	if (unlikely(i == EXIT_LOOP_COUNT))
+		panic("AMD IOMMU: Completion wait loop failed\n");
+}
+
+/*
+ * This function queues a completion wait command into the command
+ * buffer of an IOMMU
+ */
+static int __iommu_completion_wait(struct amd_iommu *iommu)
+{
+	struct iommu_cmd cmd;
+
+	 memset(&cmd, 0, sizeof(cmd));
+	 cmd.data[0] = CMD_COMPL_WAIT_INT_MASK;
+	 CMD_SET_TYPE(&cmd, CMD_COMPL_WAIT);
+
+	 return __iommu_queue_command(iommu, &cmd);
+}
+
+/*
  * This function is called whenever we need to ensure that the IOMMU has
  * completed execution of all commands we sent. It sends a
  * COMPLETION_WAIT command and waits for it to finish. The IOMMU informs
@@ -204,40 +316,22 @@ static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd)
  */
 static int iommu_completion_wait(struct amd_iommu *iommu)
 {
-	int ret = 0, ready = 0;
-	unsigned status = 0;
-	struct iommu_cmd cmd;
-	unsigned long flags, i = 0;
-
-	memset(&cmd, 0, sizeof(cmd));
-	cmd.data[0] = CMD_COMPL_WAIT_INT_MASK;
-	CMD_SET_TYPE(&cmd, CMD_COMPL_WAIT);
+	int ret = 0;
+	unsigned long flags;
 
 	spin_lock_irqsave(&iommu->lock, flags);
 
 	if (!iommu->need_sync)
 		goto out;
 
-	iommu->need_sync = 0;
+	ret = __iommu_completion_wait(iommu);
 
-	ret = __iommu_queue_command(iommu, &cmd);
+	iommu->need_sync = false;
 
 	if (ret)
 		goto out;
 
-	while (!ready && (i < EXIT_LOOP_COUNT)) {
-		++i;
-		/* wait for the bit to become one */
-		status = readl(iommu->mmio_base + MMIO_STATUS_OFFSET);
-		ready = status & MMIO_STATUS_COM_WAIT_INT_MASK;
-	}
-
-	/* set bit back to zero */
-	status &= ~MMIO_STATUS_COM_WAIT_INT_MASK;
-	writel(status, iommu->mmio_base + MMIO_STATUS_OFFSET);
-
-	if (unlikely(i == EXIT_LOOP_COUNT))
-		panic("AMD IOMMU: Completion wait loop failed\n");
+	__iommu_wait_for_completion(iommu);
 
 out:
 	spin_unlock_irqrestore(&iommu->lock, flags);
@@ -264,6 +358,21 @@ static int iommu_queue_inv_dev_entry(struct amd_iommu *iommu, u16 devid)
 	return ret;
 }
 
+static void __iommu_build_inv_iommu_pages(struct iommu_cmd *cmd, u64 address,
+					  u16 domid, int pde, int s)
+{
+	memset(cmd, 0, sizeof(*cmd));
+	address &= PAGE_MASK;
+	CMD_SET_TYPE(cmd, CMD_INV_IOMMU_PAGES);
+	cmd->data[1] |= domid;
+	cmd->data[2] = lower_32_bits(address);
+	cmd->data[3] = upper_32_bits(address);
+	if (s) /* size bit - we flush more than one 4kb page */
+		cmd->data[2] |= CMD_INV_IOMMU_PAGES_SIZE_MASK;
+	if (pde) /* PDE bit - we wan't flush everything not only the PTEs */
+		cmd->data[2] |= CMD_INV_IOMMU_PAGES_PDE_MASK;
+}
+
 /*
  * Generic command send function for invalidaing TLB entries
  */
@@ -273,16 +382,7 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu,
 	struct iommu_cmd cmd;
 	int ret;
 
-	memset(&cmd, 0, sizeof(cmd));
-	address &= PAGE_MASK;
-	CMD_SET_TYPE(&cmd, CMD_INV_IOMMU_PAGES);
-	cmd.data[1] |= domid;
-	cmd.data[2] = lower_32_bits(address);
-	cmd.data[3] = upper_32_bits(address);
-	if (s) /* size bit - we flush more than one 4kb page */
-		cmd.data[2] |= CMD_INV_IOMMU_PAGES_SIZE_MASK;
-	if (pde) /* PDE bit - we wan't flush everything not only the PTEs */
-		cmd.data[2] |= CMD_INV_IOMMU_PAGES_PDE_MASK;
+	__iommu_build_inv_iommu_pages(&cmd, address, domid, pde, s);
 
 	ret = iommu_queue_command(iommu, &cmd);
 
@@ -321,9 +421,35 @@ static void iommu_flush_tlb(struct amd_iommu *iommu, u16 domid)
 {
 	u64 address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS;
 
+	INC_STATS_COUNTER(domain_flush_single);
+
 	iommu_queue_inv_iommu_pages(iommu, address, domid, 0, 1);
 }
 
+/*
+ * This function is used to flush the IO/TLB for a given protection domain
+ * on every IOMMU in the system
+ */
+static void iommu_flush_domain(u16 domid)
+{
+	unsigned long flags;
+	struct amd_iommu *iommu;
+	struct iommu_cmd cmd;
+
+	INC_STATS_COUNTER(domain_flush_all);
+
+	__iommu_build_inv_iommu_pages(&cmd, CMD_INV_IOMMU_ALL_PAGES_ADDRESS,
+				      domid, 1, 1);
+
+	list_for_each_entry(iommu, &amd_iommu_list, list) {
+		spin_lock_irqsave(&iommu->lock, flags);
+		__iommu_queue_command(iommu, &cmd);
+		__iommu_completion_wait(iommu);
+		__iommu_wait_for_completion(iommu);
+		spin_unlock_irqrestore(&iommu->lock, flags);
+	}
+}
+
 /****************************************************************************
  *
  * The functions below are used the create the page table mappings for
@@ -338,10 +464,10 @@ static void iommu_flush_tlb(struct amd_iommu *iommu, u16 domid)
  * supporting all features of AMD IOMMU page tables like level skipping
  * and full 64 bit address spaces.
  */
-static int iommu_map(struct protection_domain *dom,
-		     unsigned long bus_addr,
-		     unsigned long phys_addr,
-		     int prot)
+static int iommu_map_page(struct protection_domain *dom,
+			  unsigned long bus_addr,
+			  unsigned long phys_addr,
+			  int prot)
 {
 	u64 __pte, *pte, *page;
 
@@ -388,6 +514,28 @@ static int iommu_map(struct protection_domain *dom,
 	return 0;
 }
 
+static void iommu_unmap_page(struct protection_domain *dom,
+			     unsigned long bus_addr)
+{
+	u64 *pte;
+
+	pte = &dom->pt_root[IOMMU_PTE_L2_INDEX(bus_addr)];
+
+	if (!IOMMU_PTE_PRESENT(*pte))
+		return;
+
+	pte = IOMMU_PTE_PAGE(*pte);
+	pte = &pte[IOMMU_PTE_L1_INDEX(bus_addr)];
+
+	if (!IOMMU_PTE_PRESENT(*pte))
+		return;
+
+	pte = IOMMU_PTE_PAGE(*pte);
+	pte = &pte[IOMMU_PTE_L1_INDEX(bus_addr)];
+
+	*pte = 0;
+}
+
 /*
  * This function checks if a specific unity mapping entry is needed for
  * this specific IOMMU.
@@ -440,7 +588,7 @@ static int dma_ops_unity_map(struct dma_ops_domain *dma_dom,
 
 	for (addr = e->address_start; addr < e->address_end;
 	     addr += PAGE_SIZE) {
-		ret = iommu_map(&dma_dom->domain, addr, addr, e->prot);
+		ret = iommu_map_page(&dma_dom->domain, addr, addr, e->prot);
 		if (ret)
 			return ret;
 		/*
@@ -571,6 +719,16 @@ static u16 domain_id_alloc(void)
 	return id;
 }
 
+static void domain_id_free(int id)
+{
+	unsigned long flags;
+
+	write_lock_irqsave(&amd_iommu_devtable_lock, flags);
+	if (id > 0 && id < MAX_DOMAIN_ID)
+		__clear_bit(id, amd_iommu_pd_alloc_bitmap);
+	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
+}
+
 /*
  * Used to reserve address ranges in the aperture (e.g. for exclusion
  * ranges.
@@ -587,12 +745,12 @@ static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 	iommu_area_reserve(dom->bitmap, start_page, pages);
 }
 
-static void dma_ops_free_pagetable(struct dma_ops_domain *dma_dom)
+static void free_pagetable(struct protection_domain *domain)
 {
 	int i, j;
 	u64 *p1, *p2, *p3;
 
-	p1 = dma_dom->domain.pt_root;
+	p1 = domain->pt_root;
 
 	if (!p1)
 		return;
@@ -613,6 +771,8 @@ static void dma_ops_free_pagetable(struct dma_ops_domain *dma_dom)
 	}
 
 	free_page((unsigned long)p1);
+
+	domain->pt_root = NULL;
 }
 
 /*
@@ -624,7 +784,7 @@ static void dma_ops_domain_free(struct dma_ops_domain *dom)
 	if (!dom)
 		return;
 
-	dma_ops_free_pagetable(dom);
+	free_pagetable(&dom->domain);
 
 	kfree(dom->pte_pages);
 
@@ -663,6 +823,7 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu,
 		goto free_dma_dom;
 	dma_dom->domain.mode = PAGE_MODE_3_LEVEL;
 	dma_dom->domain.pt_root = (void *)get_zeroed_page(GFP_KERNEL);
+	dma_dom->domain.flags = PD_DMA_OPS_MASK;
 	dma_dom->domain.priv = dma_dom;
 	if (!dma_dom->domain.pt_root)
 		goto free_dma_dom;
@@ -725,6 +886,15 @@ free_dma_dom:
 }
 
 /*
+ * little helper function to check whether a given protection domain is a
+ * dma_ops domain
+ */
+static bool dma_ops_domain(struct protection_domain *domain)
+{
+	return domain->flags & PD_DMA_OPS_MASK;
+}
+
+/*
  * Find out the protection domain structure for a given PCI device. This
  * will give us the pointer to the page table root for example.
  */
@@ -744,14 +914,15 @@ static struct protection_domain *domain_for_device(u16 devid)
  * If a device is not yet associated with a domain, this function does
  * assigns it visible for the hardware
  */
-static void set_device_domain(struct amd_iommu *iommu,
-			      struct protection_domain *domain,
-			      u16 devid)
+static void attach_device(struct amd_iommu *iommu,
+			  struct protection_domain *domain,
+			  u16 devid)
 {
 	unsigned long flags;
-
 	u64 pte_root = virt_to_phys(domain->pt_root);
 
+	domain->dev_cnt += 1;
+
 	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;
@@ -767,6 +938,116 @@ static void set_device_domain(struct amd_iommu *iommu,
 	iommu_queue_inv_dev_entry(iommu, devid);
 }
 
+/*
+ * Removes a device from a protection domain (unlocked)
+ */
+static void __detach_device(struct protection_domain *domain, u16 devid)
+{
+
+	/* lock domain */
+	spin_lock(&domain->lock);
+
+	/* remove domain from the lookup table */
+	amd_iommu_pd_table[devid] = NULL;
+
+	/* remove entry from the device table seen by the hardware */
+	amd_iommu_dev_table[devid].data[0] = IOMMU_PTE_P | IOMMU_PTE_TV;
+	amd_iommu_dev_table[devid].data[1] = 0;
+	amd_iommu_dev_table[devid].data[2] = 0;
+
+	/* decrease reference counter */
+	domain->dev_cnt -= 1;
+
+	/* ready */
+	spin_unlock(&domain->lock);
+}
+
+/*
+ * Removes a device from a protection domain (with devtable_lock held)
+ */
+static void detach_device(struct protection_domain *domain, u16 devid)
+{
+	unsigned long flags;
+
+	/* lock device table */
+	write_lock_irqsave(&amd_iommu_devtable_lock, flags);
+	__detach_device(domain, devid);
+	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
+}
+
+static int device_change_notifier(struct notifier_block *nb,
+				  unsigned long action, void *data)
+{
+	struct device *dev = data;
+	struct pci_dev *pdev = to_pci_dev(dev);
+	u16 devid = calc_devid(pdev->bus->number, pdev->devfn);
+	struct protection_domain *domain;
+	struct dma_ops_domain *dma_domain;
+	struct amd_iommu *iommu;
+	int order = amd_iommu_aperture_order;
+	unsigned long flags;
+
+	if (devid > amd_iommu_last_bdf)
+		goto out;
+
+	devid = amd_iommu_alias_table[devid];
+
+	iommu = amd_iommu_rlookup_table[devid];
+	if (iommu == NULL)
+		goto out;
+
+	domain = domain_for_device(devid);
+
+	if (domain && !dma_ops_domain(domain))
+		WARN_ONCE(1, "AMD IOMMU WARNING: device %s already bound "
+			  "to a non-dma-ops domain\n", dev_name(dev));
+
+	switch (action) {
+	case BUS_NOTIFY_BOUND_DRIVER:
+		if (domain)
+			goto out;
+		dma_domain = find_protection_domain(devid);
+		if (!dma_domain)
+			dma_domain = iommu->default_dom;
+		attach_device(iommu, &dma_domain->domain, devid);
+		printk(KERN_INFO "AMD IOMMU: Using protection domain %d for "
+		       "device %s\n", dma_domain->domain.id, dev_name(dev));
+		break;
+	case BUS_NOTIFY_UNBIND_DRIVER:
+		if (!domain)
+			goto out;
+		detach_device(domain, devid);
+		break;
+	case BUS_NOTIFY_ADD_DEVICE:
+		/* allocate a protection domain if a device is added */
+		dma_domain = find_protection_domain(devid);
+		if (dma_domain)
+			goto out;
+		dma_domain = dma_ops_domain_alloc(iommu, order);
+		if (!dma_domain)
+			goto out;
+		dma_domain->target_dev = devid;
+
+		spin_lock_irqsave(&iommu_pd_list_lock, flags);
+		list_add_tail(&dma_domain->list, &iommu_pd_list);
+		spin_unlock_irqrestore(&iommu_pd_list_lock, flags);
+
+		break;
+	default:
+		goto out;
+	}
+
+	iommu_queue_inv_dev_entry(iommu, devid);
+	iommu_completion_wait(iommu);
+
+out:
+	return 0;
+}
+
+struct notifier_block device_nb = {
+	.notifier_call = device_change_notifier,
+};
+
 /*****************************************************************************
  *
  * The next functions belong to the dma_ops mapping/unmapping code.
@@ -802,7 +1083,6 @@ static struct dma_ops_domain *find_protection_domain(u16 devid)
 	list_for_each_entry(entry, &iommu_pd_list, list) {
 		if (entry->target_dev == devid) {
 			ret = entry;
-			list_del(&ret->list);
 			break;
 		}
 	}
@@ -853,14 +1133,13 @@ static int get_device_resources(struct device *dev,
 		if (!dma_dom)
 			dma_dom = (*iommu)->default_dom;
 		*domain = &dma_dom->domain;
-		set_device_domain(*iommu, *domain, *bdf);
+		attach_device(*iommu, *domain, *bdf);
 		printk(KERN_INFO "AMD IOMMU: Using protection domain %d for "
-				"device ", (*domain)->id);
-		print_devid(_bdf, 1);
+				"device %s\n", (*domain)->id, dev_name(dev));
 	}
 
 	if (domain_for_device(_bdf) == NULL)
-		set_device_domain(*iommu, *domain, _bdf);
+		attach_device(*iommu, *domain, _bdf);
 
 	return 1;
 }
@@ -946,6 +1225,11 @@ static dma_addr_t __map_single(struct device *dev,
 	pages = iommu_num_pages(paddr, size, PAGE_SIZE);
 	paddr &= PAGE_MASK;
 
+	INC_STATS_COUNTER(total_map_requests);
+
+	if (pages > 1)
+		INC_STATS_COUNTER(cross_page);
+
 	if (align)
 		align_mask = (1UL << get_order(size)) - 1;
 
@@ -962,6 +1246,8 @@ static dma_addr_t __map_single(struct device *dev,
 	}
 	address += offset;
 
+	ADD_STATS_COUNTER(alloced_io_mem, size);
+
 	if (unlikely(dma_dom->need_flush && !amd_iommu_unmap_flush)) {
 		iommu_flush_tlb(iommu, dma_dom->domain.id);
 		dma_dom->need_flush = false;
@@ -998,6 +1284,8 @@ static void __unmap_single(struct amd_iommu *iommu,
 		start += PAGE_SIZE;
 	}
 
+	SUB_STATS_COUNTER(alloced_io_mem, size);
+
 	dma_ops_free_addresses(dma_dom, dma_addr, pages);
 
 	if (amd_iommu_unmap_flush || dma_dom->need_flush) {
@@ -1019,6 +1307,8 @@ static dma_addr_t map_single(struct device *dev, phys_addr_t paddr,
 	dma_addr_t addr;
 	u64 dma_mask;
 
+	INC_STATS_COUNTER(cnt_map_single);
+
 	if (!check_device(dev))
 		return bad_dma_address;
 
@@ -1030,6 +1320,9 @@ static dma_addr_t map_single(struct device *dev, phys_addr_t paddr,
 		/* device not handled by any AMD IOMMU */
 		return (dma_addr_t)paddr;
 
+	if (!dma_ops_domain(domain))
+		return bad_dma_address;
+
 	spin_lock_irqsave(&domain->lock, flags);
 	addr = __map_single(dev, iommu, domain->priv, paddr, size, dir, false,
 			    dma_mask);
@@ -1055,11 +1348,16 @@ static void unmap_single(struct device *dev, dma_addr_t dma_addr,
 	struct protection_domain *domain;
 	u16 devid;
 
+	INC_STATS_COUNTER(cnt_unmap_single);
+
 	if (!check_device(dev) ||
 	    !get_device_resources(dev, &iommu, &domain, &devid))
 		/* device not handled by any AMD IOMMU */
 		return;
 
+	if (!dma_ops_domain(domain))
+		return;
+
 	spin_lock_irqsave(&domain->lock, flags);
 
 	__unmap_single(iommu, domain->priv, dma_addr, size, dir);
@@ -1104,6 +1402,8 @@ static int map_sg(struct device *dev, struct scatterlist *sglist,
 	int mapped_elems = 0;
 	u64 dma_mask;
 
+	INC_STATS_COUNTER(cnt_map_sg);
+
 	if (!check_device(dev))
 		return 0;
 
@@ -1114,6 +1414,9 @@ static int map_sg(struct device *dev, struct scatterlist *sglist,
 	if (!iommu || !domain)
 		return map_sg_no_iommu(dev, sglist, nelems, dir);
 
+	if (!dma_ops_domain(domain))
+		return 0;
+
 	spin_lock_irqsave(&domain->lock, flags);
 
 	for_each_sg(sglist, s, nelems, i) {
@@ -1163,10 +1466,15 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist,
 	u16 devid;
 	int i;
 
+	INC_STATS_COUNTER(cnt_unmap_sg);
+
 	if (!check_device(dev) ||
 	    !get_device_resources(dev, &iommu, &domain, &devid))
 		return;
 
+	if (!dma_ops_domain(domain))
+		return;
+
 	spin_lock_irqsave(&domain->lock, flags);
 
 	for_each_sg(sglist, s, nelems, i) {
@@ -1194,6 +1502,8 @@ static void *alloc_coherent(struct device *dev, size_t size,
 	phys_addr_t paddr;
 	u64 dma_mask = dev->coherent_dma_mask;
 
+	INC_STATS_COUNTER(cnt_alloc_coherent);
+
 	if (!check_device(dev))
 		return NULL;
 
@@ -1212,6 +1522,9 @@ static void *alloc_coherent(struct device *dev, size_t size,
 		return virt_addr;
 	}
 
+	if (!dma_ops_domain(domain))
+		goto out_free;
+
 	if (!dma_mask)
 		dma_mask = *dev->dma_mask;
 
@@ -1220,18 +1533,20 @@ static void *alloc_coherent(struct device *dev, size_t size,
 	*dma_addr = __map_single(dev, iommu, domain->priv, paddr,
 				 size, DMA_BIDIRECTIONAL, true, dma_mask);
 
-	if (*dma_addr == bad_dma_address) {
-		free_pages((unsigned long)virt_addr, get_order(size));
-		virt_addr = NULL;
-		goto out;
-	}
+	if (*dma_addr == bad_dma_address)
+		goto out_free;
 
 	iommu_completion_wait(iommu);
 
-out:
 	spin_unlock_irqrestore(&domain->lock, flags);
 
 	return virt_addr;
+
+out_free:
+
+	free_pages((unsigned long)virt_addr, get_order(size));
+
+	return NULL;
 }
 
 /*
@@ -1245,6 +1560,8 @@ static void free_coherent(struct device *dev, size_t size,
 	struct protection_domain *domain;
 	u16 devid;
 
+	INC_STATS_COUNTER(cnt_free_coherent);
+
 	if (!check_device(dev))
 		return;
 
@@ -1253,6 +1570,9 @@ static void free_coherent(struct device *dev, size_t size,
 	if (!iommu || !domain)
 		goto free_mem;
 
+	if (!dma_ops_domain(domain))
+		goto free_mem;
+
 	spin_lock_irqsave(&domain->lock, flags);
 
 	__unmap_single(iommu, domain->priv, dma_addr, size, DMA_BIDIRECTIONAL);
@@ -1296,7 +1616,7 @@ static int amd_iommu_dma_supported(struct device *dev, u64 mask)
  * we don't need to preallocate the protection domains anymore.
  * For now we have to.
  */
-void prealloc_protection_domains(void)
+static void prealloc_protection_domains(void)
 {
 	struct pci_dev *dev = NULL;
 	struct dma_ops_domain *dma_dom;
@@ -1305,7 +1625,7 @@ void prealloc_protection_domains(void)
 	u16 devid;
 
 	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
-		devid = (dev->bus->number << 8) | dev->devfn;
+		devid = calc_devid(dev->bus->number, dev->devfn);
 		if (devid > amd_iommu_last_bdf)
 			continue;
 		devid = amd_iommu_alias_table[devid];
@@ -1352,6 +1672,7 @@ int __init amd_iommu_init_dma_ops(void)
 		iommu->default_dom = dma_ops_domain_alloc(iommu, order);
 		if (iommu->default_dom == NULL)
 			return -ENOMEM;
+		iommu->default_dom->domain.flags |= PD_DEFAULT_MASK;
 		ret = iommu_init_unity_mappings(iommu);
 		if (ret)
 			goto free_domains;
@@ -1375,6 +1696,12 @@ int __init amd_iommu_init_dma_ops(void)
 	/* Make the driver finally visible to the drivers */
 	dma_ops = &amd_iommu_dma_ops;
 
+	register_iommu(&amd_iommu_ops);
+
+	bus_register_notifier(&pci_bus_type, &device_nb);
+
+	amd_iommu_stats_init();
+
 	return 0;
 
 free_domains:
@@ -1386,3 +1713,224 @@ free_domains:
 
 	return ret;
 }
+
+/*****************************************************************************
+ *
+ * The following functions belong to the exported interface of AMD IOMMU
+ *
+ * This interface allows access to lower level functions of the IOMMU
+ * like protection domain handling and assignement of devices to domains
+ * which is not possible with the dma_ops interface.
+ *
+ *****************************************************************************/
+
+static void cleanup_domain(struct protection_domain *domain)
+{
+	unsigned long flags;
+	u16 devid;
+
+	write_lock_irqsave(&amd_iommu_devtable_lock, flags);
+
+	for (devid = 0; devid <= amd_iommu_last_bdf; ++devid)
+		if (amd_iommu_pd_table[devid] == domain)
+			__detach_device(domain, devid);
+
+	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
+}
+
+static int amd_iommu_domain_init(struct iommu_domain *dom)
+{
+	struct protection_domain *domain;
+
+	domain = kzalloc(sizeof(*domain), GFP_KERNEL);
+	if (!domain)
+		return -ENOMEM;
+
+	spin_lock_init(&domain->lock);
+	domain->mode = PAGE_MODE_3_LEVEL;
+	domain->id = domain_id_alloc();
+	if (!domain->id)
+		goto out_free;
+	domain->pt_root = (void *)get_zeroed_page(GFP_KERNEL);
+	if (!domain->pt_root)
+		goto out_free;
+
+	dom->priv = domain;
+
+	return 0;
+
+out_free:
+	kfree(domain);
+
+	return -ENOMEM;
+}
+
+static void amd_iommu_domain_destroy(struct iommu_domain *dom)
+{
+	struct protection_domain *domain = dom->priv;
+
+	if (!domain)
+		return;
+
+	if (domain->dev_cnt > 0)
+		cleanup_domain(domain);
+
+	BUG_ON(domain->dev_cnt != 0);
+
+	free_pagetable(domain);
+
+	domain_id_free(domain->id);
+
+	kfree(domain);
+
+	dom->priv = NULL;
+}
+
+static void amd_iommu_detach_device(struct iommu_domain *dom,
+				    struct device *dev)
+{
+	struct protection_domain *domain = dom->priv;
+	struct amd_iommu *iommu;
+	struct pci_dev *pdev;
+	u16 devid;
+
+	if (dev->bus != &pci_bus_type)
+		return;
+
+	pdev = to_pci_dev(dev);
+
+	devid = calc_devid(pdev->bus->number, pdev->devfn);
+
+	if (devid > 0)
+		detach_device(domain, devid);
+
+	iommu = amd_iommu_rlookup_table[devid];
+	if (!iommu)
+		return;
+
+	iommu_queue_inv_dev_entry(iommu, devid);
+	iommu_completion_wait(iommu);
+}
+
+static int amd_iommu_attach_device(struct iommu_domain *dom,
+				   struct device *dev)
+{
+	struct protection_domain *domain = dom->priv;
+	struct protection_domain *old_domain;
+	struct amd_iommu *iommu;
+	struct pci_dev *pdev;
+	u16 devid;
+
+	if (dev->bus != &pci_bus_type)
+		return -EINVAL;
+
+	pdev = to_pci_dev(dev);
+
+	devid = calc_devid(pdev->bus->number, pdev->devfn);
+
+	if (devid >= amd_iommu_last_bdf ||
+			devid != amd_iommu_alias_table[devid])
+		return -EINVAL;
+
+	iommu = amd_iommu_rlookup_table[devid];
+	if (!iommu)
+		return -EINVAL;
+
+	old_domain = domain_for_device(devid);
+	if (old_domain)
+		return -EBUSY;
+
+	attach_device(iommu, domain, devid);
+
+	iommu_completion_wait(iommu);
+
+	return 0;
+}
+
+static int amd_iommu_map_range(struct iommu_domain *dom,
+			       unsigned long iova, phys_addr_t paddr,
+			       size_t size, int iommu_prot)
+{
+	struct protection_domain *domain = dom->priv;
+	unsigned long i,  npages = iommu_num_pages(paddr, size, PAGE_SIZE);
+	int prot = 0;
+	int ret;
+
+	if (iommu_prot & IOMMU_READ)
+		prot |= IOMMU_PROT_IR;
+	if (iommu_prot & IOMMU_WRITE)
+		prot |= IOMMU_PROT_IW;
+
+	iova  &= PAGE_MASK;
+	paddr &= PAGE_MASK;
+
+	for (i = 0; i < npages; ++i) {
+		ret = iommu_map_page(domain, iova, paddr, prot);
+		if (ret)
+			return ret;
+
+		iova  += PAGE_SIZE;
+		paddr += PAGE_SIZE;
+	}
+
+	return 0;
+}
+
+static void amd_iommu_unmap_range(struct iommu_domain *dom,
+				  unsigned long iova, size_t size)
+{
+
+	struct protection_domain *domain = dom->priv;
+	unsigned long i,  npages = iommu_num_pages(iova, size, PAGE_SIZE);
+
+	iova  &= PAGE_MASK;
+
+	for (i = 0; i < npages; ++i) {
+		iommu_unmap_page(domain, iova);
+		iova  += PAGE_SIZE;
+	}
+
+	iommu_flush_domain(domain->id);
+}
+
+static phys_addr_t amd_iommu_iova_to_phys(struct iommu_domain *dom,
+					  unsigned long iova)
+{
+	struct protection_domain *domain = dom->priv;
+	unsigned long offset = iova & ~PAGE_MASK;
+	phys_addr_t paddr;
+	u64 *pte;
+
+	pte = &domain->pt_root[IOMMU_PTE_L2_INDEX(iova)];
+
+	if (!IOMMU_PTE_PRESENT(*pte))
+		return 0;
+
+	pte = IOMMU_PTE_PAGE(*pte);
+	pte = &pte[IOMMU_PTE_L1_INDEX(iova)];
+
+	if (!IOMMU_PTE_PRESENT(*pte))
+		return 0;
+
+	pte = IOMMU_PTE_PAGE(*pte);
+	pte = &pte[IOMMU_PTE_L0_INDEX(iova)];
+
+	if (!IOMMU_PTE_PRESENT(*pte))
+		return 0;
+
+	paddr  = *pte & IOMMU_PAGE_MASK;
+	paddr |= offset;
+
+	return paddr;
+}
+
+static struct iommu_ops amd_iommu_ops = {
+	.domain_init = amd_iommu_domain_init,
+	.domain_destroy = amd_iommu_domain_destroy,
+	.attach_dev = amd_iommu_attach_device,
+	.detach_dev = amd_iommu_detach_device,
+	.map = amd_iommu_map_range,
+	.unmap = amd_iommu_unmap_range,
+	.iova_to_phys = amd_iommu_iova_to_phys,
+};
+
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index c625800c55ca..42c33cebf00f 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -122,7 +122,8 @@ u16 amd_iommu_last_bdf;			/* largest PCI device id we have
 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 = 1;		/* if 1, device isolation is enabled */
+bool amd_iommu_isolate = true;		/* if true, 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
@@ -243,20 +244,16 @@ 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)
+static void __init iommu_enable(struct amd_iommu *iommu)
 {
-	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);
+	printk(KERN_INFO "AMD IOMMU: Enabling IOMMU at %s cap 0x%hx\n",
+	       dev_name(&iommu->dev->dev), 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)
+static 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);
@@ -1218,9 +1215,9 @@ static int __init parse_amd_iommu_options(char *str)
 {
 	for (; *str; ++str) {
 		if (strncmp(str, "isolate", 7) == 0)
-			amd_iommu_isolate = 1;
+			amd_iommu_isolate = true;
 		if (strncmp(str, "share", 5) == 0)
-			amd_iommu_isolate = 0;
+			amd_iommu_isolate = false;
 		if (strncmp(str, "fullflush", 9) == 0)
 			amd_iommu_unmap_flush = true;
 	}
diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c
index b5229affb953..b13d3c4dbd42 100644
--- a/arch/x86/kernel/apic.c
+++ b/arch/x86/kernel/apic.c
@@ -98,8 +98,8 @@ __setup("apicpmtimer", setup_apicpmtimer);
 #ifdef HAVE_X2APIC
 int x2apic;
 /* x2apic enabled before OS handover */
-int x2apic_preenabled;
-int disable_x2apic;
+static int x2apic_preenabled;
+static int disable_x2apic;
 static __init int setup_nox2apic(char *str)
 {
 	disable_x2apic = 1;
@@ -119,8 +119,6 @@ EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok);
 
 int first_system_vector = 0xfe;
 
-char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE};
-
 /*
  * Debug level, exported for io_apic.c
  */
@@ -142,7 +140,7 @@ static int lapic_next_event(unsigned long delta,
 			    struct clock_event_device *evt);
 static void lapic_timer_setup(enum clock_event_mode mode,
 			      struct clock_event_device *evt);
-static void lapic_timer_broadcast(cpumask_t mask);
+static void lapic_timer_broadcast(const struct cpumask *mask);
 static void apic_pm_activate(void);
 
 /*
@@ -228,7 +226,7 @@ void xapic_icr_write(u32 low, u32 id)
 	apic_write(APIC_ICR, low);
 }
 
-u64 xapic_icr_read(void)
+static u64 xapic_icr_read(void)
 {
 	u32 icr1, icr2;
 
@@ -268,7 +266,7 @@ void x2apic_icr_write(u32 low, u32 id)
 	wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low);
 }
 
-u64 x2apic_icr_read(void)
+static u64 x2apic_icr_read(void)
 {
 	unsigned long val;
 
@@ -455,7 +453,7 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 /*
  * Local APIC timer broadcast function
  */
-static void lapic_timer_broadcast(cpumask_t mask)
+static void lapic_timer_broadcast(const struct cpumask *mask)
 {
 #ifdef CONFIG_SMP
 	send_IPI_mask(mask, LOCAL_TIMER_VECTOR);
@@ -471,7 +469,7 @@ static void __cpuinit setup_APIC_timer(void)
 	struct clock_event_device *levt = &__get_cpu_var(lapic_events);
 
 	memcpy(levt, &lapic_clockevent, sizeof(*levt));
-	levt->cpumask = cpumask_of_cpu(smp_processor_id());
+	levt->cpumask = cpumask_of(smp_processor_id());
 
 	clockevents_register_device(levt);
 }
@@ -1807,28 +1805,32 @@ void disconnect_bsp_APIC(int virt_wire_setup)
 void __cpuinit generic_processor_info(int apicid, int version)
 {
 	int cpu;
-	cpumask_t tmp_map;
 
 	/*
 	 * Validate version
 	 */
 	if (version == 0x0) {
 		pr_warning("BIOS bug, APIC version is 0 for CPU#%d! "
-			"fixing up to 0x10. (tell your hw vendor)\n",
-			version);
+			   "fixing up to 0x10. (tell your hw vendor)\n",
+				version);
 		version = 0x10;
 	}
 	apic_version[apicid] = version;
 
-	if (num_processors >= NR_CPUS) {
-		pr_warning("WARNING: NR_CPUS limit of %i reached."
-			"  Processor ignored.\n", NR_CPUS);
+	if (num_processors >= nr_cpu_ids) {
+		int max = nr_cpu_ids;
+		int thiscpu = max + disabled_cpus;
+
+		pr_warning(
+			"ACPI: NR_CPUS/possible_cpus limit of %i reached."
+			"  Processor %d/0x%x ignored.\n", max, thiscpu, apicid);
+
+		disabled_cpus++;
 		return;
 	}
 
 	num_processors++;
-	cpus_complement(tmp_map, cpu_present_map);
-	cpu = first_cpu(tmp_map);
+	cpu = cpumask_next_zero(-1, cpu_present_mask);
 
 	physid_set(apicid, phys_cpu_present_map);
 	if (apicid == boot_cpu_physical_apicid) {
@@ -1878,8 +1880,8 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	}
 #endif
 
-	cpu_set(cpu, cpu_possible_map);
-	cpu_set(cpu, cpu_present_map);
+	set_cpu_possible(cpu, true);
+	set_cpu_present(cpu, true);
 }
 
 #ifdef CONFIG_X86_64
@@ -2081,7 +2083,7 @@ __cpuinit int apic_is_clustered_box(void)
 	bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid);
 	bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		/* are we being called early in kernel startup? */
 		if (bios_cpu_apicid) {
 			id = bios_cpu_apicid[i];
diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c
index 2a0a2a3cac26..f63882728d91 100644
--- a/arch/x86/kernel/bios_uv.c
+++ b/arch/x86/kernel/bios_uv.c
@@ -25,7 +25,7 @@
 #include <asm/uv/bios.h>
 #include <asm/uv/uv_hub.h>
 
-struct uv_systab uv_systab;
+static struct uv_systab uv_systab;
 
 s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5)
 {
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 42e0853030cb..3f95a40f718a 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -355,7 +355,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
 	} else if (smp_num_siblings > 1) {
 
-		if (smp_num_siblings > NR_CPUS) {
+		if (smp_num_siblings > nr_cpu_ids) {
 			printk(KERN_WARNING "CPU: Unsupported number of siblings %d",
 					smp_num_siblings);
 			smp_num_siblings = 1;
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 88ea02dcb622..28102ad1a363 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -517,6 +517,17 @@ acpi_cpufreq_guess_freq(struct acpi_cpufreq_data *data, unsigned int cpu)
 	}
 }
 
+static void free_acpi_perf_data(void)
+{
+	unsigned int i;
+
+	/* Freeing a NULL pointer is OK, and alloc_percpu zeroes. */
+	for_each_possible_cpu(i)
+		free_cpumask_var(per_cpu_ptr(acpi_perf_data, i)
+				 ->shared_cpu_map);
+	free_percpu(acpi_perf_data);
+}
+
 /*
  * acpi_cpufreq_early_init - initialize ACPI P-States library
  *
@@ -527,6 +538,7 @@ acpi_cpufreq_guess_freq(struct acpi_cpufreq_data *data, unsigned int cpu)
  */
 static int __init acpi_cpufreq_early_init(void)
 {
+	unsigned int i;
 	dprintk("acpi_cpufreq_early_init\n");
 
 	acpi_perf_data = alloc_percpu(struct acpi_processor_performance);
@@ -534,6 +546,16 @@ static int __init acpi_cpufreq_early_init(void)
 		dprintk("Memory allocation error for acpi_perf_data.\n");
 		return -ENOMEM;
 	}
+	for_each_possible_cpu(i) {
+		if (!alloc_cpumask_var_node(
+			&per_cpu_ptr(acpi_perf_data, i)->shared_cpu_map,
+			GFP_KERNEL, cpu_to_node(i))) {
+
+			/* Freeing a NULL pointer is OK: alloc_percpu zeroes. */
+			free_acpi_perf_data();
+			return -ENOMEM;
+		}
+	}
 
 	/* Do initialization in ACPI core */
 	acpi_processor_preregister_performance(acpi_perf_data);
@@ -604,9 +626,9 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	 */
 	if (policy->shared_type == CPUFREQ_SHARED_TYPE_ALL ||
 	    policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) {
-		policy->cpus = perf->shared_cpu_map;
+		cpumask_copy(&policy->cpus, perf->shared_cpu_map);
 	}
-	policy->related_cpus = perf->shared_cpu_map;
+	cpumask_copy(&policy->related_cpus, perf->shared_cpu_map);
 
 #ifdef CONFIG_SMP
 	dmi_check_system(sw_any_bug_dmi_table);
@@ -795,7 +817,7 @@ static int __init acpi_cpufreq_init(void)
 
 	ret = cpufreq_register_driver(&acpi_cpufreq_driver);
 	if (ret)
-		free_percpu(acpi_perf_data);
+		free_acpi_perf_data();
 
 	return ret;
 }
diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
index b8e05ee4f736..beea4466b063 100644
--- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
+++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
@@ -160,6 +160,7 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c)
 		switch (c->x86_model) {
 		case 0x0E: /* Core */
 		case 0x0F: /* Core Duo */
+		case 0x16: /* Celeron Core */
 			p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS;
 			return speedstep_get_processor_frequency(SPEEDSTEP_PROCESSOR_PCORE);
 		case 0x0D: /* Pentium M (Dothan) */
@@ -171,7 +172,9 @@ 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@vger.kernel.org>\n");
+		if (!cpu_has(c, X86_FEATURE_EST))
+			printk(KERN_WARNING PFX "Unknown p4-clockmod-capable CPU. "
+				"Please send an e-mail to <cpufreq@vger.kernel.org>\n");
 		return 0;
 	}
 
@@ -274,6 +277,7 @@ static struct cpufreq_driver p4clockmod_driver = {
 	.name		= "p4-clockmod",
 	.owner		= THIS_MODULE,
 	.attr		= p4clockmod_attr,
+	.hide_interface	= 1,
 };
 
 
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c
index 7c7d56b43136..1b446d79a8fd 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c
@@ -310,6 +310,12 @@ static int powernow_acpi_init(void)
 		goto err0;
 	}
 
+	if (!alloc_cpumask_var(&acpi_processor_perf->shared_cpu_map,
+								GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto err05;
+	}
+
 	if (acpi_processor_register_performance(acpi_processor_perf, 0)) {
 		retval = -EIO;
 		goto err1;
@@ -412,6 +418,8 @@ static int powernow_acpi_init(void)
 err2:
 	acpi_processor_unregister_performance(acpi_processor_perf, 0);
 err1:
+	free_cpumask_var(acpi_processor_perf->shared_cpu_map);
+err05:
 	kfree(acpi_processor_perf);
 err0:
 	printk(KERN_WARNING PFX "ACPI perflib can not be used in this platform\n");
@@ -652,6 +660,7 @@ static int powernow_cpu_exit (struct cpufreq_policy *policy) {
 #ifdef CONFIG_X86_POWERNOW_K7_ACPI
 	if (acpi_processor_perf) {
 		acpi_processor_unregister_performance(acpi_processor_perf, 0);
+		free_cpumask_var(acpi_processor_perf->shared_cpu_map);
 		kfree(acpi_processor_perf);
 	}
 #endif
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index 7f05f44b97e9..c3c9adbaa26f 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -766,7 +766,7 @@ static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned
 static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 {
 	struct cpufreq_frequency_table *powernow_table;
-	int ret_val;
+	int ret_val = -ENODEV;
 
 	if (acpi_processor_register_performance(&data->acpi_data, data->cpu)) {
 		dprintk("register performance failed: bad ACPI data\n");
@@ -815,6 +815,13 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 	/* notify BIOS that we exist */
 	acpi_processor_notify_smm(THIS_MODULE);
 
+	if (!alloc_cpumask_var(&data->acpi_data.shared_cpu_map, GFP_KERNEL)) {
+		printk(KERN_ERR PFX
+				"unable to alloc powernow_k8_data cpumask\n");
+		ret_val = -ENOMEM;
+		goto err_out_mem;
+	}
+
 	return 0;
 
 err_out_mem:
@@ -826,7 +833,7 @@ err_out:
 	/* data->acpi_data.state_count informs us at ->exit() whether ACPI was used */
 	data->acpi_data.state_count = 0;
 
-	return -ENODEV;
+	return ret_val;
 }
 
 static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table)
@@ -929,6 +936,7 @@ static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data)
 {
 	if (data->acpi_data.state_count)
 		acpi_processor_unregister_performance(&data->acpi_data, data->cpu);
+	free_cpumask_var(data->acpi_data.shared_cpu_map);
 }
 
 #else
@@ -1134,7 +1142,8 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 	data->cpu = pol->cpu;
 	data->currpstate = HW_PSTATE_INVALID;
 
-	if (powernow_k8_cpu_init_acpi(data)) {
+	rc = powernow_k8_cpu_init_acpi(data);
+	if (rc) {
 		/*
 		 * Use the PSB BIOS structure. This is only availabe on
 		 * an UP version, and is deprecated by AMD.
@@ -1152,20 +1161,17 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 			       "ACPI maintainers and complain to your BIOS "
 			       "vendor.\n");
 #endif
-			kfree(data);
-			return -ENODEV;
+			goto err_out;
 		}
 		if (pol->cpu != 0) {
 			printk(KERN_ERR FW_BUG PFX "No ACPI _PSS objects for "
 			       "CPU other than CPU0. Complain to your BIOS "
 			       "vendor.\n");
-			kfree(data);
-			return -ENODEV;
+			goto err_out;
 		}
 		rc = find_psb_table(data);
 		if (rc) {
-			kfree(data);
-			return -ENODEV;
+			goto err_out;
 		}
 	}
 
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index 3b5f06423e77..f0ea6fa2f53c 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -459,9 +459,7 @@ static int centrino_verify (struct cpufreq_policy *policy)
  * Sets a new CPUFreq policy.
  */
 struct allmasks {
-	cpumask_t		online_policy_cpus;
 	cpumask_t		saved_mask;
-	cpumask_t		set_mask;
 	cpumask_t		covered_cpus;
 };
 
@@ -475,9 +473,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 	int			retval = 0;
 	unsigned int		j, k, first_cpu, tmp;
 	CPUMASK_ALLOC(allmasks);
-	CPUMASK_PTR(online_policy_cpus, allmasks);
 	CPUMASK_PTR(saved_mask, allmasks);
-	CPUMASK_PTR(set_mask, allmasks);
 	CPUMASK_PTR(covered_cpus, allmasks);
 
 	if (unlikely(allmasks == NULL))
@@ -497,30 +493,28 @@ static int centrino_target (struct cpufreq_policy *policy,
 		goto out;
 	}
 
-#ifdef CONFIG_HOTPLUG_CPU
-	/* cpufreq holds the hotplug lock, so we are safe from here on */
-	cpus_and(*online_policy_cpus, cpu_online_map, policy->cpus);
-#else
-	*online_policy_cpus = policy->cpus;
-#endif
-
 	*saved_mask = current->cpus_allowed;
 	first_cpu = 1;
 	cpus_clear(*covered_cpus);
-	for_each_cpu_mask_nr(j, *online_policy_cpus) {
+	for_each_cpu_mask_nr(j, policy->cpus) {
+		const cpumask_t *mask;
+
+		/* cpufreq holds the hotplug lock, so we are safe here */
+		if (!cpu_online(j))
+			continue;
+
 		/*
 		 * Support for SMP systems.
 		 * Make sure we are running on CPU that wants to change freq
 		 */
-		cpus_clear(*set_mask);
 		if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY)
-			cpus_or(*set_mask, *set_mask, *online_policy_cpus);
+			mask = &policy->cpus;
 		else
-			cpu_set(j, *set_mask);
+			mask = &cpumask_of_cpu(j);
 
-		set_cpus_allowed_ptr(current, set_mask);
+		set_cpus_allowed_ptr(current, mask);
 		preempt_disable();
-		if (unlikely(!cpu_isset(smp_processor_id(), *set_mask))) {
+		if (unlikely(!cpu_isset(smp_processor_id(), *mask))) {
 			dprintk("couldn't limit to CPUs in this domain\n");
 			retval = -EAGAIN;
 			if (first_cpu) {
@@ -548,7 +542,9 @@ static int centrino_target (struct cpufreq_policy *policy,
 			dprintk("target=%dkHz old=%d new=%d msr=%04x\n",
 				target_freq, freqs.old, freqs.new, msr);
 
-			for_each_cpu_mask_nr(k, *online_policy_cpus) {
+			for_each_cpu_mask_nr(k, policy->cpus) {
+				if (!cpu_online(k))
+					continue;
 				freqs.cpu = k;
 				cpufreq_notify_transition(&freqs,
 					CPUFREQ_PRECHANGE);
@@ -571,7 +567,9 @@ static int centrino_target (struct cpufreq_policy *policy,
 		preempt_enable();
 	}
 
-	for_each_cpu_mask_nr(k, *online_policy_cpus) {
+	for_each_cpu_mask_nr(k, policy->cpus) {
+		if (!cpu_online(k))
+			continue;
 		freqs.cpu = k;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
@@ -584,18 +582,17 @@ static int centrino_target (struct cpufreq_policy *policy,
 		 * Best effort undo..
 		 */
 
-		if (!cpus_empty(*covered_cpus))
-			for_each_cpu_mask_nr(j, *covered_cpus) {
-				set_cpus_allowed_ptr(current,
-						     &cpumask_of_cpu(j));
-				wrmsr(MSR_IA32_PERF_CTL, oldmsr, h);
-			}
+		for_each_cpu_mask_nr(j, *covered_cpus) {
+			set_cpus_allowed_ptr(current, &cpumask_of_cpu(j));
+			wrmsr(MSR_IA32_PERF_CTL, oldmsr, h);
+		}
 
 		tmp = freqs.new;
 		freqs.new = freqs.old;
 		freqs.old = tmp;
-		for_each_cpu_mask_nr(j, *online_policy_cpus) {
-			freqs.cpu = j;
+		for_each_cpu_mask_nr(j, policy->cpus) {
+			if (!cpu_online(j))
+				continue;
 			cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 			cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 		}
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c
index 98d4fdb7dc04..cdac7d62369b 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c
@@ -139,6 +139,15 @@ static unsigned int pentium_core_get_frequency(void)
 	case 3:
 		fsb = 166667;
 		break;
+	case 2:
+		fsb = 200000;
+		break;
+	case 0:
+		fsb = 266667;
+		break;
+	case 4:
+		fsb = 333333;
+		break;
 	default:
 		printk(KERN_ERR "PCORE - MSR_FSB_FREQ undefined value");
 	}
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 68b5d8681cbb..48533d77be78 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -534,31 +534,16 @@ static void __cpuinit free_cache_attributes(unsigned int cpu)
 	per_cpu(cpuid4_info, cpu) = NULL;
 }
 
-static int __cpuinit detect_cache_attributes(unsigned int cpu)
+static void __cpuinit get_cpu_leaves(void *_retval)
 {
-	struct _cpuid4_info	*this_leaf;
-	unsigned long		j;
-	int			retval;
-	cpumask_t		oldmask;
-
-	if (num_cache_leaves == 0)
-		return -ENOENT;
-
-	per_cpu(cpuid4_info, cpu) = kzalloc(
-	    sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL);
-	if (per_cpu(cpuid4_info, cpu) == NULL)
-		return -ENOMEM;
-
-	oldmask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-	if (retval)
-		goto out;
+	int j, *retval = _retval, cpu = smp_processor_id();
 
 	/* Do cpuid and store the results */
 	for (j = 0; j < num_cache_leaves; j++) {
+		struct _cpuid4_info *this_leaf;
 		this_leaf = CPUID4_INFO_IDX(cpu, j);
-		retval = cpuid4_cache_lookup(j, this_leaf);
-		if (unlikely(retval < 0)) {
+		*retval = cpuid4_cache_lookup(j, this_leaf);
+		if (unlikely(*retval < 0)) {
 			int i;
 
 			for (i = 0; i < j; i++)
@@ -567,9 +552,21 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 		}
 		cache_shared_cpu_map_setup(cpu, j);
 	}
-	set_cpus_allowed_ptr(current, &oldmask);
+}
+
+static int __cpuinit detect_cache_attributes(unsigned int cpu)
+{
+	int			retval;
+
+	if (num_cache_leaves == 0)
+		return -ENOENT;
+
+	per_cpu(cpuid4_info, cpu) = kzalloc(
+	    sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL);
+	if (per_cpu(cpuid4_info, cpu) == NULL)
+		return -ENOMEM;
 
-out:
+	smp_call_function_single(cpu, get_cpu_leaves, &retval, true);
 	if (retval) {
 		kfree(per_cpu(cpuid4_info, cpu));
 		per_cpu(cpuid4_info, cpu) = NULL;
@@ -626,8 +623,8 @@ static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf,
 		cpumask_t *mask = &this_leaf->shared_cpu_map;
 
 		n = type?
-			cpulist_scnprintf(buf, len-2, *mask):
-			cpumask_scnprintf(buf, len-2, *mask);
+			cpulist_scnprintf(buf, len-2, mask) :
+			cpumask_scnprintf(buf, len-2, mask);
 		buf[n++] = '\n';
 		buf[n] = '\0';
 	}
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index 748c8f9e7a05..a5a5e0530370 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -83,34 +83,41 @@ static DEFINE_PER_CPU(unsigned char, bank_map);	/* see which banks are on */
  * CPU Initialization
  */
 
+struct thresh_restart {
+	struct threshold_block *b;
+	int reset;
+	u16 old_limit;
+};
+
 /* must be called with correct cpu affinity */
-static void threshold_restart_bank(struct threshold_block *b,
-				   int reset, u16 old_limit)
+static long threshold_restart_bank(void *_tr)
 {
+	struct thresh_restart *tr = _tr;
 	u32 mci_misc_hi, mci_misc_lo;
 
-	rdmsr(b->address, mci_misc_lo, mci_misc_hi);
+	rdmsr(tr->b->address, mci_misc_lo, mci_misc_hi);
 
-	if (b->threshold_limit < (mci_misc_hi & THRESHOLD_MAX))
-		reset = 1;	/* limit cannot be lower than err count */
+	if (tr->b->threshold_limit < (mci_misc_hi & THRESHOLD_MAX))
+		tr->reset = 1;	/* limit cannot be lower than err count */
 
-	if (reset) {		/* reset err count and overflow bit */
+	if (tr->reset) {		/* reset err count and overflow bit */
 		mci_misc_hi =
 		    (mci_misc_hi & ~(MASK_ERR_COUNT_HI | MASK_OVERFLOW_HI)) |
-		    (THRESHOLD_MAX - b->threshold_limit);
-	} else if (old_limit) {	/* change limit w/o reset */
+		    (THRESHOLD_MAX - tr->b->threshold_limit);
+	} else if (tr->old_limit) {	/* change limit w/o reset */
 		int new_count = (mci_misc_hi & THRESHOLD_MAX) +
-		    (old_limit - b->threshold_limit);
+		    (tr->old_limit - tr->b->threshold_limit);
 		mci_misc_hi = (mci_misc_hi & ~MASK_ERR_COUNT_HI) |
 		    (new_count & THRESHOLD_MAX);
 	}
 
-	b->interrupt_enable ?
+	tr->b->interrupt_enable ?
 	    (mci_misc_hi = (mci_misc_hi & ~MASK_INT_TYPE_HI) | INT_TYPE_APIC) :
 	    (mci_misc_hi &= ~MASK_INT_TYPE_HI);
 
 	mci_misc_hi |= MASK_COUNT_EN_HI;
-	wrmsr(b->address, mci_misc_lo, mci_misc_hi);
+	wrmsr(tr->b->address, mci_misc_lo, mci_misc_hi);
+	return 0;
 }
 
 /* cpu init entry point, called from mce.c with preempt off */
@@ -120,6 +127,7 @@ void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c)
 	unsigned int cpu = smp_processor_id();
 	u8 lvt_off;
 	u32 low = 0, high = 0, address = 0;
+	struct thresh_restart tr;
 
 	for (bank = 0; bank < NR_BANKS; ++bank) {
 		for (block = 0; block < NR_BLOCKS; ++block) {
@@ -162,7 +170,10 @@ void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c)
 			wrmsr(address, low, high);
 
 			threshold_defaults.address = address;
-			threshold_restart_bank(&threshold_defaults, 0, 0);
+			tr.b = &threshold_defaults;
+			tr.reset = 0;
+			tr.old_limit = 0;
+			threshold_restart_bank(&tr);
 		}
 	}
 }
@@ -251,20 +262,6 @@ struct threshold_attr {
 	ssize_t(*store) (struct threshold_block *, const char *, size_t count);
 };
 
-static void affinity_set(unsigned int cpu, cpumask_t *oldmask,
-					   cpumask_t *newmask)
-{
-	*oldmask = current->cpus_allowed;
-	cpus_clear(*newmask);
-	cpu_set(cpu, *newmask);
-	set_cpus_allowed_ptr(current, newmask);
-}
-
-static void affinity_restore(const cpumask_t *oldmask)
-{
-	set_cpus_allowed_ptr(current, oldmask);
-}
-
 #define SHOW_FIELDS(name)                                           \
 static ssize_t show_ ## name(struct threshold_block * b, char *buf) \
 {                                                                   \
@@ -277,15 +274,16 @@ static ssize_t store_interrupt_enable(struct threshold_block *b,
 				      const char *buf, size_t count)
 {
 	char *end;
-	cpumask_t oldmask, newmask;
+	struct thresh_restart tr;
 	unsigned long new = simple_strtoul(buf, &end, 0);
 	if (end == buf)
 		return -EINVAL;
 	b->interrupt_enable = !!new;
 
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 0, 0);
-	affinity_restore(&oldmask);
+	tr.b = b;
+	tr.reset = 0;
+	tr.old_limit = 0;
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 
 	return end - buf;
 }
@@ -294,8 +292,7 @@ static ssize_t store_threshold_limit(struct threshold_block *b,
 				     const char *buf, size_t count)
 {
 	char *end;
-	cpumask_t oldmask, newmask;
-	u16 old;
+	struct thresh_restart tr;
 	unsigned long new = simple_strtoul(buf, &end, 0);
 	if (end == buf)
 		return -EINVAL;
@@ -303,34 +300,36 @@ static ssize_t store_threshold_limit(struct threshold_block *b,
 		new = THRESHOLD_MAX;
 	if (new < 1)
 		new = 1;
-	old = b->threshold_limit;
+	tr.old_limit = b->threshold_limit;
 	b->threshold_limit = new;
+	tr.b = b;
+	tr.reset = 0;
 
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 0, old);
-	affinity_restore(&oldmask);
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 
 	return end - buf;
 }
 
-static ssize_t show_error_count(struct threshold_block *b, char *buf)
+static long local_error_count(void *_b)
 {
-	u32 high, low;
-	cpumask_t oldmask, newmask;
-	affinity_set(b->cpu, &oldmask, &newmask);
+	struct threshold_block *b = _b;
+	u32 low, high;
+
 	rdmsr(b->address, low, high);
-	affinity_restore(&oldmask);
-	return sprintf(buf, "%x\n",
-		       (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit));
+	return (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit);
+}
+
+static ssize_t show_error_count(struct threshold_block *b, char *buf)
+{
+	return sprintf(buf, "%lx\n", work_on_cpu(b->cpu, local_error_count, b));
 }
 
 static ssize_t store_error_count(struct threshold_block *b,
 				 const char *buf, size_t count)
 {
-	cpumask_t oldmask, newmask;
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 1, 0);
-	affinity_restore(&oldmask);
+	struct thresh_restart tr = { .b = b, .reset = 1, .old_limit = 0 };
+
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 	return 1;
 }
 
@@ -463,12 +462,19 @@ out_free:
 	return err;
 }
 
+static long local_allocate_threshold_blocks(void *_bank)
+{
+	unsigned int *bank = _bank;
+
+	return allocate_threshold_blocks(smp_processor_id(), *bank, 0,
+					 MSR_IA32_MC0_MISC + *bank * 4);
+}
+
 /* symlinks sibling shared banks to first core.  first core owns dir/files. */
 static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 {
 	int i, err = 0;
 	struct threshold_bank *b = NULL;
-	cpumask_t oldmask, newmask;
 	char name[32];
 
 	sprintf(name, "threshold_bank%i", bank);
@@ -519,11 +525,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 
 	per_cpu(threshold_banks, cpu)[bank] = b;
 
-	affinity_set(cpu, &oldmask, &newmask);
-	err = allocate_threshold_blocks(cpu, bank, 0,
-					MSR_IA32_MC0_MISC + bank * 4);
-	affinity_restore(&oldmask);
-
+	err = work_on_cpu(cpu, local_allocate_threshold_blocks, &bank);
 	if (err)
 		goto out_free;
 
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 4e8d77f01eeb..b59ddcc88cd8 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -14,14 +14,6 @@
 #include <asm/pat.h>
 #include "mtrr.h"
 
-struct mtrr_state {
-	struct mtrr_var_range var_ranges[MAX_VAR_RANGES];
-	mtrr_type fixed_ranges[NUM_FIXED_RANGES];
-	unsigned char enabled;
-	unsigned char have_fixed;
-	mtrr_type def_type;
-};
-
 struct fixed_range_block {
 	int base_msr; /* start address of an MTRR block */
 	int ranges;   /* number of MTRRs in this block  */
@@ -35,10 +27,12 @@ static struct fixed_range_block fixed_range_blocks[] = {
 };
 
 static unsigned long smp_changes_mask;
-static struct mtrr_state mtrr_state = {};
 static int mtrr_state_set;
 u64 mtrr_tom2;
 
+struct mtrr_state_type mtrr_state = {};
+EXPORT_SYMBOL_GPL(mtrr_state);
+
 #undef MODULE_PARAM_PREFIX
 #define MODULE_PARAM_PREFIX "mtrr."
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 1159e269e596..d259e5d2e054 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -49,7 +49,7 @@
 
 u32 num_var_ranges = 0;
 
-unsigned int mtrr_usage_table[MAX_VAR_RANGES];
+unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 static DEFINE_MUTEX(mtrr_mutex);
 
 u64 size_or_mask, size_and_mask;
@@ -574,7 +574,7 @@ struct mtrr_value {
 	unsigned long	lsize;
 };
 
-static struct mtrr_value mtrr_state[MAX_VAR_RANGES];
+static struct mtrr_value mtrr_state[MTRR_MAX_VAR_RANGES];
 
 static int mtrr_save(struct sys_device * sysdev, pm_message_t state)
 {
@@ -824,16 +824,14 @@ static int enable_mtrr_cleanup __initdata =
 
 static int __init disable_mtrr_cleanup_setup(char *str)
 {
-	if (enable_mtrr_cleanup != -1)
-		enable_mtrr_cleanup = 0;
+	enable_mtrr_cleanup = 0;
 	return 0;
 }
 early_param("disable_mtrr_cleanup", disable_mtrr_cleanup_setup);
 
 static int __init enable_mtrr_cleanup_setup(char *str)
 {
-	if (enable_mtrr_cleanup != -1)
-		enable_mtrr_cleanup = 1;
+	enable_mtrr_cleanup = 1;
 	return 0;
 }
 early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup);
diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index 2dc4ec656b23..ffd60409cc6d 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -8,11 +8,6 @@
 #define MTRRcap_MSR     0x0fe
 #define MTRRdefType_MSR 0x2ff
 
-#define MTRRphysBase_MSR(reg) (0x200 + 2 * (reg))
-#define MTRRphysMask_MSR(reg) (0x200 + 2 * (reg) + 1)
-
-#define NUM_FIXED_RANGES 88
-#define MAX_VAR_RANGES 256
 #define MTRRfix64K_00000_MSR 0x250
 #define MTRRfix16K_80000_MSR 0x258
 #define MTRRfix16K_A0000_MSR 0x259
@@ -29,11 +24,7 @@
 #define MTRR_CHANGE_MASK_VARIABLE  0x02
 #define MTRR_CHANGE_MASK_DEFTYPE   0x04
 
-/* In the Intel processor's MTRR interface, the MTRR type is always held in
-   an 8 bit field: */
-typedef u8 mtrr_type;
-
-extern unsigned int mtrr_usage_table[MAX_VAR_RANGES];
+extern unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 
 struct mtrr_ops {
 	u32	vendor;
@@ -70,13 +61,6 @@ struct set_mtrr_context {
 	u32 ccr3;
 };
 
-struct mtrr_var_range {
-	u32 base_lo;
-	u32 base_hi;
-	u32 mask_lo;
-	u32 mask_hi;
-};
-
 void set_mtrr_done(struct set_mtrr_context *ctxt);
 void set_mtrr_cache_disable(struct set_mtrr_context *ctxt);
 void set_mtrr_prepare_save(struct set_mtrr_context *ctxt);
diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c
index 72cefd1e649b..2ac1f0c2beb3 100644
--- a/arch/x86/kernel/cpuid.c
+++ b/arch/x86/kernel/cpuid.c
@@ -39,10 +39,10 @@
 #include <linux/device.h>
 #include <linux/cpu.h>
 #include <linux/notifier.h>
+#include <linux/uaccess.h>
 
 #include <asm/processor.h>
 #include <asm/msr.h>
-#include <asm/uaccess.h>
 #include <asm/system.h>
 
 static struct class *cpuid_class;
@@ -82,7 +82,7 @@ static loff_t cpuid_seek(struct file *file, loff_t offset, int orig)
 }
 
 static ssize_t cpuid_read(struct file *file, char __user *buf,
-			  size_t count, loff_t * ppos)
+			  size_t count, loff_t *ppos)
 {
 	char __user *tmp = buf;
 	struct cpuid_regs cmd;
@@ -117,11 +117,11 @@ static int cpuid_open(struct inode *inode, struct file *file)
 	unsigned int cpu;
 	struct cpuinfo_x86 *c;
 	int ret = 0;
-	
+
 	lock_kernel();
 
 	cpu = iminor(file->f_path.dentry->d_inode);
-	if (cpu >= NR_CPUS || !cpu_online(cpu)) {
+	if (cpu >= nr_cpu_ids || !cpu_online(cpu)) {
 		ret = -ENXIO;	/* No such CPU */
 		goto out;
 	}
diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index d84a852e4cd7..c689d19e35ab 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -26,6 +26,7 @@
 #include <linux/kdebug.h>
 #include <asm/smp.h>
 #include <asm/reboot.h>
+#include <asm/virtext.h>
 
 #include <mach_ipi.h>
 
@@ -49,6 +50,15 @@ static void kdump_nmi_callback(int cpu, struct die_args *args)
 #endif
 	crash_save_cpu(regs, cpu);
 
+	/* Disable VMX or SVM if needed.
+	 *
+	 * We need to disable virtualization on all CPUs.
+	 * Having VMX or SVM enabled on any CPU may break rebooting
+	 * after the kdump kernel has finished its task.
+	 */
+	cpu_emergency_vmxoff();
+	cpu_emergency_svm_disable();
+
 	disable_local_APIC();
 }
 
@@ -80,6 +90,14 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
 	local_irq_disable();
 
 	kdump_nmi_shootdown_cpus();
+
+	/* Booting kdump kernel with VMX or SVM enabled won't work,
+	 * because (among other limitations) we can't disable paging
+	 * with the virt flags.
+	 */
+	cpu_emergency_vmxoff();
+	cpu_emergency_svm_disable();
+
 	lapic_shutdown();
 #if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index 23b138e31e9c..504ad198e4ad 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -886,7 +886,7 @@ asmlinkage void early_printk(const char *fmt, ...)
 	va_list ap;
 
 	va_start(ap, fmt);
-	n = vscnprintf(buf, 512, fmt, ap);
+	n = vscnprintf(buf, sizeof(buf), fmt, ap);
 	early_console->write(early_console, buf, n);
 	va_end(ap);
 }
diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c
index c0262791bda4..34185488e4fb 100644
--- a/arch/x86/kernel/genapic_flat_64.c
+++ b/arch/x86/kernel/genapic_flat_64.c
@@ -30,12 +30,12 @@ static int flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 1;
 }
 
-static cpumask_t flat_target_cpus(void)
+static const struct cpumask *flat_target_cpus(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 
-static cpumask_t flat_vector_allocation_domain(int cpu)
+static void flat_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -45,8 +45,8 @@ static cpumask_t flat_vector_allocation_domain(int cpu)
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	cpumask_t domain = { { [0] = APIC_ALL_CPUS, } };
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
 }
 
 /*
@@ -69,9 +69,8 @@ static void flat_init_apic_ldr(void)
 	apic_write(APIC_LDR, val);
 }
 
-static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
+static inline void _flat_send_IPI_mask(unsigned long mask, int vector)
 {
-	unsigned long mask = cpus_addr(cpumask)[0];
 	unsigned long flags;
 
 	local_irq_save(flags);
@@ -79,20 +78,41 @@ static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
 	local_irq_restore(flags);
 }
 
+static void flat_send_IPI_mask(const struct cpumask *cpumask, int vector)
+{
+	unsigned long mask = cpumask_bits(cpumask)[0];
+
+	_flat_send_IPI_mask(mask, vector);
+}
+
+static void flat_send_IPI_mask_allbutself(const struct cpumask *cpumask,
+					  int vector)
+{
+	unsigned long mask = cpumask_bits(cpumask)[0];
+	int cpu = smp_processor_id();
+
+	if (cpu < BITS_PER_LONG)
+		clear_bit(cpu, &mask);
+	_flat_send_IPI_mask(mask, vector);
+}
+
 static void flat_send_IPI_allbutself(int vector)
 {
+	int cpu = smp_processor_id();
 #ifdef	CONFIG_HOTPLUG_CPU
 	int hotplug = 1;
 #else
 	int hotplug = 0;
 #endif
 	if (hotplug || vector == NMI_VECTOR) {
-		cpumask_t allbutme = cpu_online_map;
+		if (!cpumask_equal(cpu_online_mask, cpumask_of(cpu))) {
+			unsigned long mask = cpumask_bits(cpu_online_mask)[0];
 
-		cpu_clear(smp_processor_id(), allbutme);
+			if (cpu < BITS_PER_LONG)
+				clear_bit(cpu, &mask);
 
-		if (!cpus_empty(allbutme))
-			flat_send_IPI_mask(allbutme, vector);
+			_flat_send_IPI_mask(mask, vector);
+		}
 	} else if (num_online_cpus() > 1) {
 		__send_IPI_shortcut(APIC_DEST_ALLBUT, vector,APIC_DEST_LOGICAL);
 	}
@@ -101,7 +121,7 @@ static void flat_send_IPI_allbutself(int vector)
 static void flat_send_IPI_all(int vector)
 {
 	if (vector == NMI_VECTOR)
-		flat_send_IPI_mask(cpu_online_map, vector);
+		flat_send_IPI_mask(cpu_online_mask, vector);
 	else
 		__send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
 }
@@ -135,9 +155,18 @@ static int flat_apic_id_registered(void)
 	return physid_isset(read_xapic_id(), phys_cpu_present_map);
 }
 
-static unsigned int flat_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int flat_cpu_mask_to_apicid(const struct cpumask *cpumask)
+{
+	return cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
+}
+
+static unsigned int flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						const struct cpumask *andmask)
 {
-	return cpus_addr(cpumask)[0] & APIC_ALL_CPUS;
+	unsigned long mask1 = cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
+	unsigned long mask2 = cpumask_bits(andmask)[0] & APIC_ALL_CPUS;
+
+	return mask1 & mask2;
 }
 
 static unsigned int phys_pkg_id(int index_msb)
@@ -157,8 +186,10 @@ struct genapic apic_flat =  {
 	.send_IPI_all = flat_send_IPI_all,
 	.send_IPI_allbutself = flat_send_IPI_allbutself,
 	.send_IPI_mask = flat_send_IPI_mask,
+	.send_IPI_mask_allbutself = flat_send_IPI_mask_allbutself,
 	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = flat_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = flat_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
@@ -188,35 +219,39 @@ static int physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 0;
 }
 
-static cpumask_t physflat_target_cpus(void)
+static const struct cpumask *physflat_target_cpus(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 
-static cpumask_t physflat_vector_allocation_domain(int cpu)
+static void physflat_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	return cpumask_of_cpu(cpu);
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
-static void physflat_send_IPI_mask(cpumask_t cpumask, int vector)
+static void physflat_send_IPI_mask(const struct cpumask *cpumask, int vector)
 {
 	send_IPI_mask_sequence(cpumask, vector);
 }
 
-static void physflat_send_IPI_allbutself(int vector)
+static void physflat_send_IPI_mask_allbutself(const struct cpumask *cpumask,
+					      int vector)
 {
-	cpumask_t allbutme = cpu_online_map;
+	send_IPI_mask_allbutself(cpumask, vector);
+}
 
-	cpu_clear(smp_processor_id(), allbutme);
-	physflat_send_IPI_mask(allbutme, vector);
+static void physflat_send_IPI_allbutself(int vector)
+{
+	send_IPI_mask_allbutself(cpu_online_mask, vector);
 }
 
 static void physflat_send_IPI_all(int vector)
 {
-	physflat_send_IPI_mask(cpu_online_map, vector);
+	physflat_send_IPI_mask(cpu_online_mask, vector);
 }
 
-static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int physflat_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -224,13 +259,31 @@ static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = cpumask_first(cpumask);
 	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int
+physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+				const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 struct genapic apic_physflat =  {
 	.name = "physical flat",
 	.acpi_madt_oem_check = physflat_acpi_madt_oem_check,
@@ -243,8 +296,10 @@ struct genapic apic_physflat =  {
 	.send_IPI_all = physflat_send_IPI_all,
 	.send_IPI_allbutself = physflat_send_IPI_allbutself,
 	.send_IPI_mask = physflat_send_IPI_mask,
+	.send_IPI_mask_allbutself = physflat_send_IPI_mask_allbutself,
 	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = physflat_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = physflat_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c
index f6a2c8eb48a6..6ce497cc372d 100644
--- a/arch/x86/kernel/genx2apic_cluster.c
+++ b/arch/x86/kernel/genx2apic_cluster.c
@@ -22,19 +22,18 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t x2apic_target_cpus(void)
+static const struct cpumask *x2apic_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
 /*
  * for now each logical cpu is in its own vector allocation domain.
  */
-static cpumask_t x2apic_vector_allocation_domain(int cpu)
+static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
@@ -56,32 +55,53 @@ static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
  * at once. We have 16 cpu's in a cluster. This will minimize IPI register
  * writes.
  */
-static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned long query_cpu;
 
 	local_irq_save(flags);
-	for_each_cpu_mask(query_cpu, mask) {
-		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_logical_apicid, query_cpu),
-				       vector, APIC_DEST_LOGICAL);
-	}
+	for_each_cpu(query_cpu, mask)
+		__x2apic_send_IPI_dest(
+			per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+			vector, APIC_DEST_LOGICAL);
 	local_irq_restore(flags);
 }
 
-static void x2apic_send_IPI_allbutself(int vector)
+static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask,
+					    int vector)
 {
-	cpumask_t mask = cpu_online_map;
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	cpu_clear(smp_processor_id(), mask);
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+				vector, APIC_DEST_LOGICAL);
+	local_irq_restore(flags);
+}
+
+static void x2apic_send_IPI_allbutself(int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	if (!cpus_empty(mask))
-		x2apic_send_IPI_mask(mask, vector);
+	local_irq_save(flags);
+	for_each_online_cpu(query_cpu)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+				vector, APIC_DEST_LOGICAL);
+	local_irq_restore(flags);
 }
 
 static void x2apic_send_IPI_all(int vector)
 {
-	x2apic_send_IPI_mask(cpu_online_map, vector);
+	x2apic_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int x2apic_apic_id_registered(void)
@@ -89,21 +109,38 @@ static int x2apic_apic_id_registered(void)
 	return 1;
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
 	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	cpu = cpumask_first(cpumask);
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_logical_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_logical_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -150,8 +187,10 @@ struct genapic apic_x2apic_cluster = {
 	.send_IPI_all = x2apic_send_IPI_all,
 	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
 	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself,
 	.send_IPI_self = x2apic_send_IPI_self,
 	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c
index d042211768b7..21bcc0e098ba 100644
--- a/arch/x86/kernel/genx2apic_phys.c
+++ b/arch/x86/kernel/genx2apic_phys.c
@@ -29,16 +29,15 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t x2apic_target_cpus(void)
+static const struct cpumask *x2apic_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
-static cpumask_t x2apic_vector_allocation_domain(int cpu)
+static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
@@ -54,32 +53,54 @@ static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
 	x2apic_icr_write(cfg, apicid);
 }
 
-static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned long query_cpu;
 
 	local_irq_save(flags);
-	for_each_cpu_mask(query_cpu, mask) {
+	for_each_cpu(query_cpu, mask) {
 		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu),
 				       vector, APIC_DEST_PHYSICAL);
 	}
 	local_irq_restore(flags);
 }
 
-static void x2apic_send_IPI_allbutself(int vector)
+static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask,
+					    int vector)
 {
-	cpumask_t mask = cpu_online_map;
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask) {
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_apicid, query_cpu),
+				vector, APIC_DEST_PHYSICAL);
+	}
+	local_irq_restore(flags);
+}
 
-	cpu_clear(smp_processor_id(), mask);
+static void x2apic_send_IPI_allbutself(int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	if (!cpus_empty(mask))
-		x2apic_send_IPI_mask(mask, vector);
+	local_irq_save(flags);
+	for_each_online_cpu(query_cpu)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_apicid, query_cpu),
+				vector, APIC_DEST_PHYSICAL);
+	local_irq_restore(flags);
 }
 
 static void x2apic_send_IPI_all(int vector)
 {
-	x2apic_send_IPI_mask(cpu_online_map, vector);
+	x2apic_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int x2apic_apic_id_registered(void)
@@ -87,7 +108,7 @@ static int x2apic_apic_id_registered(void)
 	return 1;
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -95,13 +116,30 @@ static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	cpu = cpumask_first(cpumask);
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -123,12 +161,12 @@ static unsigned int phys_pkg_id(int index_msb)
 	return current_cpu_data.initial_apicid >> index_msb;
 }
 
-void x2apic_send_IPI_self(int vector)
+static void x2apic_send_IPI_self(int vector)
 {
 	apic_write(APIC_SELF_IPI, vector);
 }
 
-void init_x2apic_ldr(void)
+static void init_x2apic_ldr(void)
 {
 	return;
 }
@@ -145,8 +183,10 @@ struct genapic apic_x2apic_phys = {
 	.send_IPI_all = x2apic_send_IPI_all,
 	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
 	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself,
 	.send_IPI_self = x2apic_send_IPI_self,
 	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
index dece17289731..b193e082f6ce 100644
--- a/arch/x86/kernel/genx2apic_uv_x.c
+++ b/arch/x86/kernel/genx2apic_uv_x.c
@@ -79,16 +79,15 @@ EXPORT_SYMBOL(sn_rtc_cycles_per_second);
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t uv_target_cpus(void)
+static const struct cpumask *uv_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
-static cpumask_t uv_vector_allocation_domain(int cpu)
+static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
@@ -127,28 +126,37 @@ static void uv_send_IPI_one(int cpu, int vector)
 	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
 }
 
-static void uv_send_IPI_mask(cpumask_t mask, int vector)
+static void uv_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned int cpu;
 
-	for_each_possible_cpu(cpu)
-		if (cpu_isset(cpu, mask))
+	for_each_cpu(cpu, mask)
+		uv_send_IPI_one(cpu, vector);
+}
+
+static void uv_send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+	unsigned int cpu;
+	unsigned int this_cpu = smp_processor_id();
+
+	for_each_cpu(cpu, mask)
+		if (cpu != this_cpu)
 			uv_send_IPI_one(cpu, vector);
 }
 
 static void uv_send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
-
-	cpu_clear(smp_processor_id(), mask);
+	unsigned int cpu;
+	unsigned int this_cpu = smp_processor_id();
 
-	if (!cpus_empty(mask))
-		uv_send_IPI_mask(mask, vector);
+	for_each_online_cpu(cpu)
+		if (cpu != this_cpu)
+			uv_send_IPI_one(cpu, vector);
 }
 
 static void uv_send_IPI_all(int vector)
 {
-	uv_send_IPI_mask(cpu_online_map, vector);
+	uv_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int uv_apic_id_registered(void)
@@ -160,7 +168,7 @@ static void uv_init_apic_ldr(void)
 {
 }
 
-static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int uv_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -168,13 +176,30 @@ static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = cpumask_first(cpumask);
 	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+					      const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -222,8 +247,10 @@ struct genapic apic_x2apic_uv_x = {
 	.send_IPI_all = uv_send_IPI_all,
 	.send_IPI_allbutself = uv_send_IPI_allbutself,
 	.send_IPI_mask = uv_send_IPI_mask,
+	.send_IPI_mask_allbutself = uv_send_IPI_mask_allbutself,
 	.send_IPI_self = uv_send_IPI_self,
 	.cpu_mask_to_apicid = uv_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = uv_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 388e05a5fc17..b9a4d8c4b935 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -27,7 +27,7 @@
 #include <asm/trampoline.h>
 
 /* boot cpu pda */
-static struct x8664_pda _boot_cpu_pda __read_mostly;
+static struct x8664_pda _boot_cpu_pda;
 
 #ifdef CONFIG_SMP
 /*
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 845ea097383e..cd759ad90690 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -248,7 +248,7 @@ static void hpet_legacy_clockevent_register(void)
 	 * Start hpet with the boot cpu mask and make it
 	 * global after the IO_APIC has been initialized.
 	 */
-	hpet_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	hpet_clockevent.cpumask = cpumask_of(smp_processor_id());
 	clockevents_register_device(&hpet_clockevent);
 	global_clock_event = &hpet_clockevent;
 	printk(KERN_DEBUG "hpet clockevent registered\n");
@@ -303,7 +303,7 @@ static void hpet_set_mode(enum clock_event_mode mode,
 			struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
 			hpet_setup_msi_irq(hdev->irq);
 			disable_irq(hdev->irq);
-			irq_set_affinity(hdev->irq, cpumask_of_cpu(hdev->cpu));
+			irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu));
 			enable_irq(hdev->irq);
 		}
 		break;
@@ -451,7 +451,7 @@ static int hpet_setup_irq(struct hpet_dev *dev)
 		return -1;
 
 	disable_irq(dev->irq);
-	irq_set_affinity(dev->irq, cpumask_of_cpu(dev->cpu));
+	irq_set_affinity(dev->irq, cpumask_of(dev->cpu));
 	enable_irq(dev->irq);
 
 	printk(KERN_DEBUG "hpet: %s irq %d for MSI\n",
@@ -502,7 +502,7 @@ static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu)
 	/* 5 usec minimum reprogramming delta. */
 	evt->min_delta_ns = 5000;
 
-	evt->cpumask = cpumask_of_cpu(hdev->cpu);
+	evt->cpumask = cpumask_of(hdev->cpu);
 	clockevents_register_device(evt);
 }
 
diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c
index c1b5e3ece1f2..10f92fb532f3 100644
--- a/arch/x86/kernel/i8253.c
+++ b/arch/x86/kernel/i8253.c
@@ -114,7 +114,7 @@ void __init setup_pit_timer(void)
 	 * Start pit with the boot cpu mask and make it global after the
 	 * IO_APIC has been initialized.
 	 */
-	pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	pit_clockevent.cpumask = cpumask_of(smp_processor_id());
 	pit_clockevent.mult = div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC,
 				     pit_clockevent.shift);
 	pit_clockevent.max_delta_ns =
diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index 74917658b004..3639442aa7a4 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -136,8 +136,8 @@ static struct irq_pin_list *get_one_free_irq_2_pin(int cpu)
 
 struct irq_cfg {
 	struct irq_pin_list *irq_2_pin;
-	cpumask_t domain;
-	cpumask_t old_domain;
+	cpumask_var_t domain;
+	cpumask_var_t old_domain;
 	unsigned move_cleanup_count;
 	u8 vector;
 	u8 move_in_progress : 1;
@@ -152,22 +152,22 @@ static struct irq_cfg irq_cfgx[] = {
 #else
 static struct irq_cfg irq_cfgx[NR_IRQS] = {
 #endif
-	[0]  = { .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR,  },
-	[1]  = { .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR,  },
-	[2]  = { .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR,  },
-	[3]  = { .domain = CPU_MASK_ALL, .vector = IRQ3_VECTOR,  },
-	[4]  = { .domain = CPU_MASK_ALL, .vector = IRQ4_VECTOR,  },
-	[5]  = { .domain = CPU_MASK_ALL, .vector = IRQ5_VECTOR,  },
-	[6]  = { .domain = CPU_MASK_ALL, .vector = IRQ6_VECTOR,  },
-	[7]  = { .domain = CPU_MASK_ALL, .vector = IRQ7_VECTOR,  },
-	[8]  = { .domain = CPU_MASK_ALL, .vector = IRQ8_VECTOR,  },
-	[9]  = { .domain = CPU_MASK_ALL, .vector = IRQ9_VECTOR,  },
-	[10] = { .domain = CPU_MASK_ALL, .vector = IRQ10_VECTOR, },
-	[11] = { .domain = CPU_MASK_ALL, .vector = IRQ11_VECTOR, },
-	[12] = { .domain = CPU_MASK_ALL, .vector = IRQ12_VECTOR, },
-	[13] = { .domain = CPU_MASK_ALL, .vector = IRQ13_VECTOR, },
-	[14] = { .domain = CPU_MASK_ALL, .vector = IRQ14_VECTOR, },
-	[15] = { .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, },
+	[0]  = { .vector = IRQ0_VECTOR,  },
+	[1]  = { .vector = IRQ1_VECTOR,  },
+	[2]  = { .vector = IRQ2_VECTOR,  },
+	[3]  = { .vector = IRQ3_VECTOR,  },
+	[4]  = { .vector = IRQ4_VECTOR,  },
+	[5]  = { .vector = IRQ5_VECTOR,  },
+	[6]  = { .vector = IRQ6_VECTOR,  },
+	[7]  = { .vector = IRQ7_VECTOR,  },
+	[8]  = { .vector = IRQ8_VECTOR,  },
+	[9]  = { .vector = IRQ9_VECTOR,  },
+	[10] = { .vector = IRQ10_VECTOR, },
+	[11] = { .vector = IRQ11_VECTOR, },
+	[12] = { .vector = IRQ12_VECTOR, },
+	[13] = { .vector = IRQ13_VECTOR, },
+	[14] = { .vector = IRQ14_VECTOR, },
+	[15] = { .vector = IRQ15_VECTOR, },
 };
 
 int __init arch_early_irq_init(void)
@@ -183,6 +183,10 @@ int __init arch_early_irq_init(void)
 	for (i = 0; i < count; i++) {
 		desc = irq_to_desc(i);
 		desc->chip_data = &cfg[i];
+		alloc_bootmem_cpumask_var(&cfg[i].domain);
+		alloc_bootmem_cpumask_var(&cfg[i].old_domain);
+		if (i < NR_IRQS_LEGACY)
+			cpumask_setall(cfg[i].domain);
 	}
 
 	return 0;
@@ -209,6 +213,20 @@ static struct irq_cfg *get_one_free_irq_cfg(int cpu)
 	node = cpu_to_node(cpu);
 
 	cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node);
+	if (cfg) {
+		if (!alloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) {
+			kfree(cfg);
+			cfg = NULL;
+		} else if (!alloc_cpumask_var_node(&cfg->old_domain,
+							  GFP_ATOMIC, node)) {
+			free_cpumask_var(cfg->domain);
+			kfree(cfg);
+			cfg = NULL;
+		} else {
+			cpumask_clear(cfg->domain);
+			cpumask_clear(cfg->old_domain);
+		}
+	}
 	printk(KERN_DEBUG "  alloc irq_cfg on cpu %d node %d\n", cpu, node);
 
 	return cfg;
@@ -333,13 +351,14 @@ void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc)
 	}
 }
 
-static void set_extra_move_desc(struct irq_desc *desc, cpumask_t mask)
+static void
+set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg = desc->chip_data;
 
 	if (!cfg->move_in_progress) {
 		/* it means that domain is not changed */
-		if (!cpus_intersects(desc->affinity, mask))
+		if (!cpumask_intersects(&desc->affinity, mask))
 			cfg->move_desc_pending = 1;
 	}
 }
@@ -354,7 +373,8 @@ static struct irq_cfg *irq_cfg(unsigned int irq)
 #endif
 
 #ifndef CONFIG_NUMA_MIGRATE_IRQ_DESC
-static inline void set_extra_move_desc(struct irq_desc *desc, cpumask_t mask)
+static inline void
+set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 }
 #endif
@@ -485,6 +505,26 @@ static void ioapic_mask_entry(int apic, int pin)
 }
 
 #ifdef CONFIG_SMP
+static void send_cleanup_vector(struct irq_cfg *cfg)
+{
+	cpumask_var_t cleanup_mask;
+
+	if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) {
+		unsigned int i;
+		cfg->move_cleanup_count = 0;
+		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
+			cfg->move_cleanup_count++;
+		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
+			send_IPI_mask(cpumask_of(i), IRQ_MOVE_CLEANUP_VECTOR);
+	} else {
+		cpumask_and(cleanup_mask, cfg->old_domain, cpu_online_mask);
+		cfg->move_cleanup_count = cpumask_weight(cleanup_mask);
+		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
+		free_cpumask_var(cleanup_mask);
+	}
+	cfg->move_in_progress = 0;
+}
+
 static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg)
 {
 	int apic, pin;
@@ -520,41 +560,55 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq
 	}
 }
 
-static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask);
+static int
+assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask);
 
-static void set_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mask)
+/*
+ * Either sets desc->affinity to a valid value, and returns cpu_mask_to_apicid
+ * of that, or returns BAD_APICID and leaves desc->affinity untouched.
+ */
+static unsigned int
+set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
-	unsigned long flags;
-	unsigned int dest;
-	cpumask_t tmp;
 	unsigned int irq;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
-		return;
+	if (!cpumask_intersects(mask, cpu_online_mask))
+		return BAD_APICID;
 
 	irq = desc->irq;
 	cfg = desc->chip_data;
 	if (assign_irq_vector(irq, cfg, mask))
-		return;
+		return BAD_APICID;
 
+	cpumask_and(&desc->affinity, cfg->domain, mask);
 	set_extra_move_desc(desc, mask);
+	return cpu_mask_to_apicid_and(&desc->affinity, cpu_online_mask);
+}
 
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
-	/*
-	 * Only the high 8 bits are valid.
-	 */
-	dest = SET_APIC_LOGICAL_ID(dest);
+static void
+set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
+{
+	struct irq_cfg *cfg;
+	unsigned long flags;
+	unsigned int dest;
+	unsigned int irq;
+
+	irq = desc->irq;
+	cfg = desc->chip_data;
 
 	spin_lock_irqsave(&ioapic_lock, flags);
-	__target_IO_APIC_irq(irq, dest, cfg);
-	desc->affinity = mask;
+	dest = set_desc_affinity(desc, mask);
+	if (dest != BAD_APICID) {
+		/* Only the high 8 bits are valid. */
+		dest = SET_APIC_LOGICAL_ID(dest);
+		__target_IO_APIC_irq(irq, dest, cfg);
+	}
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+static void
+set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc;
 
@@ -652,7 +706,7 @@ static void __unmask_IO_APIC_irq(struct irq_cfg *cfg)
 }
 
 #ifdef CONFIG_X86_64
-void io_apic_sync(struct irq_pin_list *entry)
+static void io_apic_sync(struct irq_pin_list *entry)
 {
 	/*
 	 * Synchronize the IO-APIC and the CPU by doing
@@ -1222,7 +1276,8 @@ void unlock_vector_lock(void)
 	spin_unlock(&vector_lock);
 }
 
-static int __assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
+static int
+__assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 {
 	/*
 	 * NOTE! The local APIC isn't very good at handling
@@ -1237,49 +1292,49 @@ static int __assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
 	 */
 	static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0;
 	unsigned int old_vector;
-	int cpu;
+	int cpu, err;
+	cpumask_var_t tmp_mask;
 
 	if ((cfg->move_in_progress) || cfg->move_cleanup_count)
 		return -EBUSY;
 
-	/* Only try and allocate irqs on cpus that are present */
-	cpus_and(mask, mask, cpu_online_map);
+	if (!alloc_cpumask_var(&tmp_mask, GFP_ATOMIC))
+		return -ENOMEM;
 
 	old_vector = cfg->vector;
 	if (old_vector) {
-		cpumask_t tmp;
-		cpus_and(tmp, cfg->domain, mask);
-		if (!cpus_empty(tmp))
+		cpumask_and(tmp_mask, mask, cpu_online_mask);
+		cpumask_and(tmp_mask, cfg->domain, tmp_mask);
+		if (!cpumask_empty(tmp_mask)) {
+			free_cpumask_var(tmp_mask);
 			return 0;
+		}
 	}
 
-	for_each_cpu_mask_nr(cpu, mask) {
-		cpumask_t domain, new_mask;
+	/* Only try and allocate irqs on cpus that are present */
+	err = -ENOSPC;
+	for_each_cpu_and(cpu, mask, cpu_online_mask) {
 		int new_cpu;
 		int vector, offset;
 
-		domain = vector_allocation_domain(cpu);
-		cpus_and(new_mask, domain, cpu_online_map);
+		vector_allocation_domain(cpu, tmp_mask);
 
 		vector = current_vector;
 		offset = current_offset;
 next:
 		vector += 8;
 		if (vector >= first_system_vector) {
-			/* If we run out of vectors on large boxen, must share them. */
+			/* If out of vectors on large boxen, must share them. */
 			offset = (offset + 1) % 8;
 			vector = FIRST_DEVICE_VECTOR + offset;
 		}
 		if (unlikely(current_vector == vector))
 			continue;
-#ifdef CONFIG_X86_64
-		if (vector == IA32_SYSCALL_VECTOR)
-			goto next;
-#else
-		if (vector == SYSCALL_VECTOR)
+
+		if (test_bit(vector, used_vectors))
 			goto next;
-#endif
-		for_each_cpu_mask_nr(new_cpu, new_mask)
+
+		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask)
 			if (per_cpu(vector_irq, new_cpu)[vector] != -1)
 				goto next;
 		/* Found one! */
@@ -1287,18 +1342,21 @@ next:
 		current_offset = offset;
 		if (old_vector) {
 			cfg->move_in_progress = 1;
-			cfg->old_domain = cfg->domain;
+			cpumask_copy(cfg->old_domain, cfg->domain);
 		}
-		for_each_cpu_mask_nr(new_cpu, new_mask)
+		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask)
 			per_cpu(vector_irq, new_cpu)[vector] = irq;
 		cfg->vector = vector;
-		cfg->domain = domain;
-		return 0;
+		cpumask_copy(cfg->domain, tmp_mask);
+		err = 0;
+		break;
 	}
-	return -ENOSPC;
+	free_cpumask_var(tmp_mask);
+	return err;
 }
 
-static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
+static int
+assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 {
 	int err;
 	unsigned long flags;
@@ -1311,23 +1369,20 @@ static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
 
 static void __clear_irq_vector(int irq, struct irq_cfg *cfg)
 {
-	cpumask_t mask;
 	int cpu, vector;
 
 	BUG_ON(!cfg->vector);
 
 	vector = cfg->vector;
-	cpus_and(mask, cfg->domain, cpu_online_map);
-	for_each_cpu_mask_nr(cpu, mask)
+	for_each_cpu_and(cpu, cfg->domain, cpu_online_mask)
 		per_cpu(vector_irq, cpu)[vector] = -1;
 
 	cfg->vector = 0;
-	cpus_clear(cfg->domain);
+	cpumask_clear(cfg->domain);
 
 	if (likely(!cfg->move_in_progress))
 		return;
-	cpus_and(mask, cfg->old_domain, cpu_online_map);
-	for_each_cpu_mask_nr(cpu, mask) {
+	for_each_cpu_and(cpu, cfg->old_domain, cpu_online_mask) {
 		for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS;
 								vector++) {
 			if (per_cpu(vector_irq, cpu)[vector] != irq)
@@ -1350,7 +1405,7 @@ void __setup_vector_irq(int cpu)
 	/* Mark the inuse vectors */
 	for_each_irq_desc(irq, desc) {
 		cfg = desc->chip_data;
-		if (!cpu_isset(cpu, cfg->domain))
+		if (!cpumask_test_cpu(cpu, cfg->domain))
 			continue;
 		vector = cfg->vector;
 		per_cpu(vector_irq, cpu)[vector] = irq;
@@ -1362,7 +1417,7 @@ void __setup_vector_irq(int cpu)
 			continue;
 
 		cfg = irq_cfg(irq);
-		if (!cpu_isset(cpu, cfg->domain))
+		if (!cpumask_test_cpu(cpu, cfg->domain))
 			per_cpu(vector_irq, cpu)[vector] = -1;
 	}
 }
@@ -1498,18 +1553,17 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_de
 {
 	struct irq_cfg *cfg;
 	struct IO_APIC_route_entry entry;
-	cpumask_t mask;
+	unsigned int dest;
 
 	if (!IO_APIC_IRQ(irq))
 		return;
 
 	cfg = desc->chip_data;
 
-	mask = TARGET_CPUS;
-	if (assign_irq_vector(irq, cfg, mask))
+	if (assign_irq_vector(irq, cfg, TARGET_CPUS))
 		return;
 
-	cpus_and(mask, cfg->domain, mask);
+	dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 	apic_printk(APIC_VERBOSE,KERN_DEBUG
 		    "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> "
@@ -1519,8 +1573,7 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_de
 
 
 	if (setup_ioapic_entry(mp_ioapics[apic].mp_apicid, irq, &entry,
-			       cpu_mask_to_apicid(mask), trigger, polarity,
-			       cfg->vector)) {
+			       dest, trigger, polarity, cfg->vector)) {
 		printk("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
 		       mp_ioapics[apic].mp_apicid, pin);
 		__clear_irq_vector(irq, cfg);
@@ -2240,7 +2293,7 @@ static int ioapic_retrigger_irq(unsigned int irq)
 	unsigned long flags;
 
 	spin_lock_irqsave(&vector_lock, flags);
-	send_IPI_mask(cpumask_of_cpu(first_cpu(cfg->domain)), cfg->vector);
+	send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector);
 	spin_unlock_irqrestore(&vector_lock, flags);
 
 	return 1;
@@ -2289,18 +2342,17 @@ static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration);
  * as simple as edge triggered migration and we can do the irq migration
  * with a simple atomic update to IO-APIC RTE.
  */
-static void migrate_ioapic_irq_desc(struct irq_desc *desc, cpumask_t mask)
+static void
+migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
-	cpumask_t tmp, cleanup_mask;
 	struct irte irte;
 	int modify_ioapic_rte;
 	unsigned int dest;
 	unsigned long flags;
 	unsigned int irq;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
 	irq = desc->irq;
@@ -2313,8 +2365,7 @@ static void migrate_ioapic_irq_desc(struct irq_desc *desc, cpumask_t mask)
 
 	set_extra_move_desc(desc, mask);
 
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
+	dest = cpu_mask_to_apicid_and(cfg->domain, mask);
 
 	modify_ioapic_rte = desc->status & IRQ_LEVEL;
 	if (modify_ioapic_rte) {
@@ -2331,14 +2382,10 @@ static void migrate_ioapic_irq_desc(struct irq_desc *desc, cpumask_t mask)
 	 */
 	modify_irte(irq, &irte);
 
-	if (cfg->move_in_progress) {
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
+	if (cfg->move_in_progress)
+		send_cleanup_vector(cfg);
 
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 
 static int migrate_irq_remapped_level_desc(struct irq_desc *desc)
@@ -2360,11 +2407,11 @@ static int migrate_irq_remapped_level_desc(struct irq_desc *desc)
 	}
 
 	/* everthing is clear. we have right of way */
-	migrate_ioapic_irq_desc(desc, desc->pending_mask);
+	migrate_ioapic_irq_desc(desc, &desc->pending_mask);
 
 	ret = 0;
 	desc->status &= ~IRQ_MOVE_PENDING;
-	cpus_clear(desc->pending_mask);
+	cpumask_clear(&desc->pending_mask);
 
 unmask:
 	unmask_IO_APIC_irq_desc(desc);
@@ -2389,7 +2436,7 @@ static void ir_irq_migration(struct work_struct *work)
 				continue;
 			}
 
-			desc->chip->set_affinity(irq, desc->pending_mask);
+			desc->chip->set_affinity(irq, &desc->pending_mask);
 			spin_unlock_irqrestore(&desc->lock, flags);
 		}
 	}
@@ -2398,18 +2445,20 @@ static void ir_irq_migration(struct work_struct *work)
 /*
  * Migrates the IRQ destination in the process context.
  */
-static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mask)
+static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
+					    const struct cpumask *mask)
 {
 	if (desc->status & IRQ_LEVEL) {
 		desc->status |= IRQ_MOVE_PENDING;
-		desc->pending_mask = mask;
+		cpumask_copy(&desc->pending_mask, mask);
 		migrate_irq_remapped_level_desc(desc);
 		return;
 	}
 
 	migrate_ioapic_irq_desc(desc, mask);
 }
-static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+static void set_ir_ioapic_affinity_irq(unsigned int irq,
+				       const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
@@ -2444,7 +2493,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
 		if (!cfg->move_cleanup_count)
 			goto unlock;
 
-		if ((vector == cfg->vector) && cpu_isset(me, cfg->domain))
+		if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
 			goto unlock;
 
 		__get_cpu_var(vector_irq)[vector] = -1;
@@ -2481,20 +2530,14 @@ static void irq_complete_move(struct irq_desc **descp)
 
 	vector = ~get_irq_regs()->orig_ax;
 	me = smp_processor_id();
-	if ((vector == cfg->vector) && cpu_isset(me, cfg->domain)) {
-		cpumask_t cleanup_mask;
-
 #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
 		*descp = desc = move_irq_desc(desc, me);
 		/* get the new one */
 		cfg = desc->chip_data;
 #endif
 
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
+	if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
+		send_cleanup_vector(cfg);
 }
 #else
 static inline void irq_complete_move(struct irq_desc **descp) {}
@@ -3216,16 +3259,13 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 	struct irq_cfg *cfg;
 	int err;
 	unsigned dest;
-	cpumask_t tmp;
 
 	cfg = irq_cfg(irq);
-	tmp = TARGET_CPUS;
-	err = assign_irq_vector(irq, cfg, tmp);
+	err = assign_irq_vector(irq, cfg, TARGET_CPUS);
 	if (err)
 		return err;
 
-	cpus_and(tmp, cfg->domain, tmp);
-	dest = cpu_mask_to_apicid(tmp);
+	dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 #ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
@@ -3279,26 +3319,18 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 }
 
 #ifdef CONFIG_SMP
-static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	read_msi_msg_desc(desc, &msg);
 
@@ -3308,37 +3340,27 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	write_msi_msg_desc(desc, &msg);
-	desc->affinity = mask;
 }
 #ifdef CONFIG_INTR_REMAP
 /*
  * Migrate the MSI irq to another cpumask. This migration is
  * done in the process context using interrupt-remapping hardware.
  */
-static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void
+ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg;
+	struct irq_cfg *cfg = desc->chip_data;
 	unsigned int dest;
-	cpumask_t tmp, cleanup_mask;
 	struct irte irte;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
-		return;
-
 	if (get_irte(irq, &irte))
 		return;
 
-	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
-
 	irte.vector = cfg->vector;
 	irte.dest_id = IRTE_DEST(dest);
 
@@ -3352,14 +3374,8 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 	 * at the new destination. So, time to cleanup the previous
 	 * vector allocation.
 	 */
-	if (cfg->move_in_progress) {
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
-
-	desc->affinity = mask;
+	if (cfg->move_in_progress)
+		send_cleanup_vector(cfg);
 }
 
 #endif
@@ -3550,26 +3566,18 @@ void arch_teardown_msi_irq(unsigned int irq)
 
 #ifdef CONFIG_DMAR
 #ifdef CONFIG_SMP
-static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	dmar_msi_read(irq, &msg);
 
@@ -3579,7 +3587,6 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	dmar_msi_write(irq, &msg);
-	desc->affinity = mask;
 }
 
 #endif /* CONFIG_SMP */
@@ -3613,26 +3620,18 @@ int arch_setup_dmar_msi(unsigned int irq)
 #ifdef CONFIG_HPET_TIMER
 
 #ifdef CONFIG_SMP
-static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	hpet_msi_read(irq, &msg);
 
@@ -3642,7 +3641,6 @@ static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	hpet_msi_write(irq, &msg);
-	desc->affinity = mask;
 }
 
 #endif /* CONFIG_SMP */
@@ -3697,28 +3695,19 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector)
 	write_ht_irq_msg(irq, &msg);
 }
 
-static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	target_ht_irq(irq, dest, cfg->vector);
-	desc->affinity = mask;
 }
 
 #endif
@@ -3738,17 +3727,14 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
 {
 	struct irq_cfg *cfg;
 	int err;
-	cpumask_t tmp;
 
 	cfg = irq_cfg(irq);
-	tmp = TARGET_CPUS;
-	err = assign_irq_vector(irq, cfg, tmp);
+	err = assign_irq_vector(irq, cfg, TARGET_CPUS);
 	if (!err) {
 		struct ht_irq_msg msg;
 		unsigned dest;
 
-		cpus_and(tmp, cfg->domain, tmp);
-		dest = cpu_mask_to_apicid(tmp);
+		dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 		msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest);
 
@@ -3784,7 +3770,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
 int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 		       unsigned long mmr_offset)
 {
-	const cpumask_t *eligible_cpu = get_cpu_mask(cpu);
+	const struct cpumask *eligible_cpu = cpumask_of(cpu);
 	struct irq_cfg *cfg;
 	int mmr_pnode;
 	unsigned long mmr_value;
@@ -3794,7 +3780,7 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 
 	cfg = irq_cfg(irq);
 
-	err = assign_irq_vector(irq, cfg, *eligible_cpu);
+	err = assign_irq_vector(irq, cfg, eligible_cpu);
 	if (err != 0)
 		return err;
 
@@ -3813,7 +3799,7 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 	entry->polarity = 0;
 	entry->trigger = 0;
 	entry->mask = 0;
-	entry->dest = cpu_mask_to_apicid(*eligible_cpu);
+	entry->dest = cpu_mask_to_apicid(eligible_cpu);
 
 	mmr_pnode = uv_blade_to_pnode(mmr_blade);
 	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
@@ -4024,7 +4010,7 @@ void __init setup_ioapic_dest(void)
 	int pin, ioapic, irq, irq_entry;
 	struct irq_desc *desc;
 	struct irq_cfg *cfg;
-	cpumask_t mask;
+	const struct cpumask *mask;
 
 	if (skip_ioapic_setup == 1)
 		return;
@@ -4055,7 +4041,7 @@ void __init setup_ioapic_dest(void)
 			 */
 			if (desc->status &
 			    (IRQ_NO_BALANCING | IRQ_AFFINITY_SET))
-				mask = desc->affinity;
+				mask = &desc->affinity;
 			else
 				mask = TARGET_CPUS;
 
diff --git a/arch/x86/kernel/ipi.c b/arch/x86/kernel/ipi.c
index f1c688e46f35..285bbf8831fa 100644
--- a/arch/x86/kernel/ipi.c
+++ b/arch/x86/kernel/ipi.c
@@ -116,18 +116,18 @@ static inline void __send_IPI_dest_field(unsigned long mask, int vector)
 /*
  * This is only used on smaller machines.
  */
-void send_IPI_mask_bitmask(cpumask_t cpumask, int vector)
+void send_IPI_mask_bitmask(const struct cpumask *cpumask, int vector)
 {
-	unsigned long mask = cpus_addr(cpumask)[0];
+	unsigned long mask = cpumask_bits(cpumask)[0];
 	unsigned long flags;
 
 	local_irq_save(flags);
-	WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]);
+	WARN_ON(mask & ~cpumask_bits(cpu_online_mask)[0]);
 	__send_IPI_dest_field(mask, vector);
 	local_irq_restore(flags);
 }
 
-void send_IPI_mask_sequence(cpumask_t mask, int vector)
+void send_IPI_mask_sequence(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned int query_cpu;
@@ -139,12 +139,24 @@ void send_IPI_mask_sequence(cpumask_t mask, int vector)
 	 */
 
 	local_irq_save(flags);
-	for_each_possible_cpu(query_cpu) {
-		if (cpu_isset(query_cpu, mask)) {
+	for_each_cpu(query_cpu, mask)
+		__send_IPI_dest_field(cpu_to_logical_apicid(query_cpu), vector);
+	local_irq_restore(flags);
+}
+
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+	unsigned long flags;
+	unsigned int query_cpu;
+	unsigned int this_cpu = smp_processor_id();
+
+	/* See Hack comment above */
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask)
+		if (query_cpu != this_cpu)
 			__send_IPI_dest_field(cpu_to_logical_apicid(query_cpu),
 					      vector);
-		}
-	}
 	local_irq_restore(flags);
 }
 
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 3f1d9d18df67..bce53e1352a0 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -9,6 +9,7 @@
 #include <asm/apic.h>
 #include <asm/io_apic.h>
 #include <asm/smp.h>
+#include <asm/irq.h>
 
 atomic_t irq_err_count;
 
@@ -190,3 +191,5 @@ u64 arch_irq_stat(void)
 #endif
 	return sum;
 }
+
+EXPORT_SYMBOL_GPL(vector_used_by_percpu_irq);
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 119fc9c8ff7f..9dc5588f336a 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -233,27 +233,28 @@ unsigned int do_IRQ(struct pt_regs *regs)
 #ifdef CONFIG_HOTPLUG_CPU
 #include <mach_apic.h>
 
-void fixup_irqs(cpumask_t map)
+/* A cpu has been removed from cpu_online_mask.  Reset irq affinities. */
+void fixup_irqs(void)
 {
 	unsigned int irq;
 	static int warned;
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		cpumask_t mask;
+		const struct cpumask *affinity;
 
 		if (!desc)
 			continue;
 		if (irq == 2)
 			continue;
 
-		cpus_and(mask, desc->affinity, map);
-		if (any_online_cpu(mask) == NR_CPUS) {
+		affinity = &desc->affinity;
+		if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
 			printk("Breaking affinity for irq %i\n", irq);
-			mask = map;
+			affinity = cpu_all_mask;
 		}
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, affinity);
 		else if (desc->action && !(warned++))
 			printk("Cannot set affinity for irq %i\n", irq);
 	}
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index a174a217eb1a..6383d50f82ea 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -80,16 +80,17 @@ asmlinkage unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-void fixup_irqs(cpumask_t map)
+/* A cpu has been removed from cpu_online_mask.  Reset irq affinities. */
+void fixup_irqs(void)
 {
 	unsigned int irq;
 	static int warned;
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		cpumask_t mask;
 		int break_affinity = 0;
 		int set_affinity = 1;
+		const struct cpumask *affinity;
 
 		if (!desc)
 			continue;
@@ -99,23 +100,23 @@ void fixup_irqs(cpumask_t map)
 		/* interrupt's are disabled at this point */
 		spin_lock(&desc->lock);
 
+		affinity = &desc->affinity;
 		if (!irq_has_action(irq) ||
-		    cpus_equal(desc->affinity, map)) {
+		    cpumask_equal(affinity, cpu_online_mask)) {
 			spin_unlock(&desc->lock);
 			continue;
 		}
 
-		cpus_and(mask, desc->affinity, map);
-		if (cpus_empty(mask)) {
+		if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
 			break_affinity = 1;
-			mask = map;
+			affinity = cpu_all_mask;
 		}
 
 		if (desc->chip->mask)
 			desc->chip->mask(irq);
 
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, affinity);
 		else if (!(warned++))
 			set_affinity = 0;
 
diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c
index 203384ed2b5d..84723295f88a 100644
--- a/arch/x86/kernel/irqinit_32.c
+++ b/arch/x86/kernel/irqinit_32.c
@@ -110,6 +110,18 @@ DEFINE_PER_CPU(vector_irq_t, vector_irq) = {
 	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
 };
 
+int vector_used_by_percpu_irq(unsigned int vector)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		if (per_cpu(vector_irq, cpu)[vector] != -1)
+			return 1;
+	}
+
+	return 0;
+}
+
 /* Overridden in paravirt.c */
 void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ")));
 
@@ -146,10 +158,12 @@ void __init native_init_IRQ(void)
 	alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
 
 	/* IPI for single call function */
-	set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, call_function_single_interrupt);
+	alloc_intr_gate(CALL_FUNCTION_SINGLE_VECTOR,
+				 call_function_single_interrupt);
 
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
+	set_bit(IRQ_MOVE_CLEANUP_VECTOR, used_vectors);
 #endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 6190e6ef546c..31ebfe38e96c 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -69,6 +69,18 @@ DEFINE_PER_CPU(vector_irq_t, vector_irq) = {
 	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
 };
 
+int vector_used_by_percpu_irq(unsigned int vector)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		if (per_cpu(vector_irq, cpu)[vector] != -1)
+			return 1;
+	}
+
+	return 0;
+}
+
 void __init init_ISA_irqs(void)
 {
 	int i;
@@ -121,6 +133,7 @@ static void __init smp_intr_init(void)
 
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
+	set_bit(IRQ_MOVE_CLEANUP_VECTOR, used_vectors);
 #endif
 }
 
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 6c27679ec6aa..884d985b8b82 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -376,9 +376,10 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
-	mutex_unlock(&kprobe_mutex);
+	if (p->ainsn.insn) {
+		free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
+		p->ainsn.insn = NULL;
+	}
 }
 
 static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
@@ -694,7 +695,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 	/*
 	 * It is possible to have multiple instances associated with a given
 	 * task either because multiple functions in the call path have
-	 * return probes installed on them, and/or more then one
+	 * return probes installed on them, and/or more than one
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index e169ae9b6a62..652fce6d2cce 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -89,17 +89,17 @@ static cycle_t kvm_clock_read(void)
  */
 static unsigned long kvm_get_tsc_khz(void)
 {
-	return preset_lpj;
+	struct pvclock_vcpu_time_info *src;
+	src = &per_cpu(hv_clock, 0);
+	return pvclock_tsc_khz(src);
 }
 
 static void kvm_get_preset_lpj(void)
 {
-	struct pvclock_vcpu_time_info *src;
 	unsigned long khz;
 	u64 lpj;
 
-	src = &per_cpu(hv_clock, 0);
-	khz = pvclock_tsc_khz(src);
+	khz = kvm_get_tsc_khz();
 
 	lpj = ((u64)khz * 1000);
 	do_div(lpj, HZ);
@@ -194,5 +194,7 @@ void __init kvmclock_init(void)
 #endif
 		kvm_get_preset_lpj();
 		clocksource_register(&kvm_clock);
+		pv_info.paravirt_enabled = 1;
+		pv_info.name = "KVM";
 	}
 }
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index eee32b43fee3..71f1d99a635d 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -12,8 +12,8 @@
 #include <linux/mm.h>
 #include <linux/smp.h>
 #include <linux/vmalloc.h>
+#include <linux/uaccess.h>
 
-#include <asm/uaccess.h>
 #include <asm/system.h>
 #include <asm/ldt.h>
 #include <asm/desc.h>
@@ -93,7 +93,7 @@ static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
 	if (err < 0)
 		return err;
 
-	for(i = 0; i < old->size; i++)
+	for (i = 0; i < old->size; i++)
 		write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE);
 	return 0;
 }
diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c
index 3b599518c322..8815f3c7fec7 100644
--- a/arch/x86/kernel/mfgpt_32.c
+++ b/arch/x86/kernel/mfgpt_32.c
@@ -252,7 +252,7 @@ EXPORT_SYMBOL_GPL(geode_mfgpt_alloc_timer);
 /*
  * The MFPGT timers on the CS5536 provide us with suitable timers to use
  * as clock event sources - not as good as a HPET or APIC, but certainly
- * better then the PIT.  This isn't a general purpose MFGPT driver, but
+ * better than the PIT.  This isn't a general purpose MFGPT driver, but
  * a simplified one designed specifically to act as a clock event source.
  * For full details about the MFGPT, please consult the CS5536 data sheet.
  */
@@ -287,7 +287,7 @@ static struct clock_event_device mfgpt_clockevent = {
 	.set_mode = mfgpt_set_mode,
 	.set_next_event = mfgpt_next_event,
 	.rating = 250,
-	.cpumask = CPU_MASK_ALL,
+	.cpumask = cpu_all_mask,
 	.shift = 32
 };
 
diff --git a/arch/x86/kernel/mmconf-fam10h_64.c b/arch/x86/kernel/mmconf-fam10h_64.c
index efc2f361fe85..666e43df51f9 100644
--- a/arch/x86/kernel/mmconf-fam10h_64.c
+++ b/arch/x86/kernel/mmconf-fam10h_64.c
@@ -13,8 +13,7 @@
 #include <asm/msr.h>
 #include <asm/acpi.h>
 #include <asm/mmconfig.h>
-
-#include "../pci/pci.h"
+#include <asm/pci_x86.h>
 
 struct pci_hostbridge_probe {
 	u32 bus;
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 45e3b69808ba..c5c5b8df1dbc 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -16,14 +16,14 @@
 #include <linux/bitops.h>
 #include <linux/acpi.h>
 #include <linux/module.h>
+#include <linux/smp.h>
+#include <linux/acpi.h>
 
-#include <asm/smp.h>
 #include <asm/mtrr.h>
 #include <asm/mpspec.h>
 #include <asm/pgalloc.h>
 #include <asm/io_apic.h>
 #include <asm/proto.h>
-#include <asm/acpi.h>
 #include <asm/bios_ebda.h>
 #include <asm/e820.h>
 #include <asm/trampoline.h>
@@ -95,8 +95,8 @@ static void __init MP_bus_info(struct mpc_config_bus *m)
 #endif
 
 	if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA) - 1) == 0) {
-		 set_bit(m->mpc_busid, mp_bus_not_pci);
-#if defined(CONFIG_EISA) || defined (CONFIG_MCA)
+		set_bit(m->mpc_busid, mp_bus_not_pci);
+#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
 #endif
 	} else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI) - 1) == 0) {
@@ -104,7 +104,7 @@ static void __init MP_bus_info(struct mpc_config_bus *m)
 			x86_quirks->mpc_oem_pci_bus(m);
 
 		clear_bit(m->mpc_busid, mp_bus_not_pci);
-#if defined(CONFIG_EISA) || defined (CONFIG_MCA)
+#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
 	} else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA) - 1) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c
index 82a7c7ed6d45..726266695b2c 100644
--- a/arch/x86/kernel/msr.c
+++ b/arch/x86/kernel/msr.c
@@ -136,7 +136,7 @@ static int msr_open(struct inode *inode, struct file *file)
 	lock_kernel();
 	cpu = iminor(file->f_path.dentry->d_inode);
 
-	if (cpu >= NR_CPUS || !cpu_online(cpu)) {
+	if (cpu >= nr_cpu_ids || !cpu_online(cpu)) {
 		ret = -ENXIO;	/* No such CPU */
 		goto out;
 	}
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 8bd1bf9622a7..45a09ccdc214 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -26,11 +26,10 @@
 #include <linux/kernel_stat.h>
 #include <linux/kdebug.h>
 #include <linux/smp.h>
+#include <linux/nmi.h>
 
 #include <asm/i8259.h>
 #include <asm/io_apic.h>
-#include <asm/smp.h>
-#include <asm/nmi.h>
 #include <asm/proto.h>
 #include <asm/timer.h>
 
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 19a1044a0cd9..b25428533141 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -38,7 +38,7 @@ EXPORT_SYMBOL(bad_dma_address);
    be probably a smaller DMA mask, but this is bug-to-bug compatible
    to older i386. */
 struct device x86_dma_fallback_dev = {
-	.bus_id = "fallback device",
+	.init_name = "fallback device",
 	.coherent_dma_mask = DMA_32BIT_MASK,
 	.dma_mask = &x86_dma_fallback_dev.coherent_dma_mask,
 };
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index a35eaa379ff6..00c2bcd41463 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -52,7 +52,7 @@ static u32 *iommu_gatt_base;		/* Remapping table */
  * to trigger bugs with some popular PCI cards, in particular 3ware (but
  * has been also also seen with Qlogic at least).
  */
-int iommu_fullflush = 1;
+static int iommu_fullflush = 1;
 
 /* Allocation bitmap for the remapping area: */
 static DEFINE_SPINLOCK(iommu_bitmap_lock);
diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index 242c3440687f..d59c91747665 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -13,7 +13,7 @@
 
 int swiotlb __read_mostly;
 
-void *swiotlb_alloc_boot(size_t size, unsigned long nslabs)
+void * __init swiotlb_alloc_boot(size_t size, unsigned long nslabs)
 {
 	return alloc_bootmem_low_pages(size);
 }
@@ -23,7 +23,7 @@ void *swiotlb_alloc(unsigned order, unsigned long nslabs)
 	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
 }
 
-dma_addr_t swiotlb_phys_to_bus(phys_addr_t paddr)
+dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr)
 {
 	return paddr;
 }
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 61f718df6eec..2b46eb41643b 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -12,6 +12,8 @@
 #include <asm/proto.h>
 #include <asm/reboot_fixups.h>
 #include <asm/reboot.h>
+#include <asm/pci_x86.h>
+#include <asm/virtext.h>
 
 #ifdef CONFIG_X86_32
 # include <linux/dmi.h>
@@ -23,7 +25,6 @@
 
 #include <mach_ipi.h>
 
-
 /*
  * Power off function, if any
  */
@@ -39,6 +40,12 @@ int reboot_force;
 static int reboot_cpu = -1;
 #endif
 
+/* This is set if we need to go through the 'emergency' path.
+ * When machine_emergency_restart() is called, we may be on
+ * an inconsistent state and won't be able to do a clean cleanup
+ */
+static int reboot_emergency;
+
 /* This is set by the PCI code if either type 1 or type 2 PCI is detected */
 bool port_cf9_safe = false;
 
@@ -368,6 +375,48 @@ static inline void kb_wait(void)
 	}
 }
 
+static void vmxoff_nmi(int cpu, struct die_args *args)
+{
+	cpu_emergency_vmxoff();
+}
+
+/* Use NMIs as IPIs to tell all CPUs to disable virtualization
+ */
+static void emergency_vmx_disable_all(void)
+{
+	/* Just make sure we won't change CPUs while doing this */
+	local_irq_disable();
+
+	/* We need to disable VMX on all CPUs before rebooting, otherwise
+	 * we risk hanging up the machine, because the CPU ignore INIT
+	 * signals when VMX is enabled.
+	 *
+	 * We can't take any locks and we may be on an inconsistent
+	 * state, so we use NMIs as IPIs to tell the other CPUs to disable
+	 * VMX and halt.
+	 *
+	 * For safety, we will avoid running the nmi_shootdown_cpus()
+	 * stuff unnecessarily, but we don't have a way to check
+	 * if other CPUs have VMX enabled. So we will call it only if the
+	 * CPU we are running on has VMX enabled.
+	 *
+	 * We will miss cases where VMX is not enabled on all CPUs. This
+	 * shouldn't do much harm because KVM always enable VMX on all
+	 * CPUs anyway. But we can miss it on the small window where KVM
+	 * is still enabling VMX.
+	 */
+	if (cpu_has_vmx() && cpu_vmx_enabled()) {
+		/* Disable VMX on this CPU.
+		 */
+		cpu_vmxoff();
+
+		/* Halt and disable VMX on the other CPUs */
+		nmi_shootdown_cpus(vmxoff_nmi);
+
+	}
+}
+
+
 void __attribute__((weak)) mach_reboot_fixups(void)
 {
 }
@@ -376,6 +425,9 @@ static void native_machine_emergency_restart(void)
 {
 	int i;
 
+	if (reboot_emergency)
+		emergency_vmx_disable_all();
+
 	/* Tell the BIOS if we want cold or warm reboot */
 	*((unsigned short *)__va(0x472)) = reboot_mode;
 
@@ -449,7 +501,7 @@ void native_machine_shutdown(void)
 
 #ifdef CONFIG_X86_32
 	/* See if there has been given a command line override */
-	if ((reboot_cpu != -1) && (reboot_cpu < NR_CPUS) &&
+	if ((reboot_cpu != -1) && (reboot_cpu < nr_cpu_ids) &&
 		cpu_online(reboot_cpu))
 		reboot_cpu_id = reboot_cpu;
 #endif
@@ -459,7 +511,7 @@ void native_machine_shutdown(void)
 		reboot_cpu_id = smp_processor_id();
 
 	/* Make certain I only run on the appropriate processor */
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(reboot_cpu_id));
+	set_cpus_allowed_ptr(current, cpumask_of(reboot_cpu_id));
 
 	/* O.K Now that I'm on the appropriate processor,
 	 * stop all of the others.
@@ -482,13 +534,19 @@ void native_machine_shutdown(void)
 #endif
 }
 
+static void __machine_emergency_restart(int emergency)
+{
+	reboot_emergency = emergency;
+	machine_ops.emergency_restart();
+}
+
 static void native_machine_restart(char *__unused)
 {
 	printk("machine restart\n");
 
 	if (!reboot_force)
 		machine_shutdown();
-	machine_emergency_restart();
+	__machine_emergency_restart(0);
 }
 
 static void native_machine_halt(void)
@@ -532,7 +590,7 @@ void machine_shutdown(void)
 
 void machine_emergency_restart(void)
 {
-	machine_ops.emergency_restart();
+	__machine_emergency_restart(1);
 }
 
 void machine_restart(char *cmd)
@@ -592,10 +650,7 @@ static int crash_nmi_callback(struct notifier_block *self,
 
 static void smp_send_nmi_allbutself(void)
 {
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(safe_smp_processor_id(), mask);
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, NMI_VECTOR);
+	send_IPI_allbutself(NMI_VECTOR);
 }
 
 static struct notifier_block crash_nmi_nb = {
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index ae0c0d3bb770..a4b619c33106 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -152,8 +152,11 @@ void __init setup_per_cpu_areas(void)
 	old_size = PERCPU_ENOUGH_ROOM;
 	align = max_t(unsigned long, PAGE_SIZE, align);
 	size = roundup(old_size, align);
-	printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
-			  size);
+
+	pr_info("NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n",
+		NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids);
+
+	pr_info("PERCPU: Allocating %zd bytes of per cpu data\n", size);
 
 	for_each_possible_cpu(cpu) {
 #ifndef CONFIG_NEED_MULTIPLE_NODES
@@ -164,28 +167,21 @@ void __init setup_per_cpu_areas(void)
 		if (!node_online(node) || !NODE_DATA(node)) {
 			ptr = __alloc_bootmem(size, align,
 					 __pa(MAX_DMA_ADDRESS));
-			printk(KERN_INFO
-			       "cpu %d has no node %d or node-local memory\n",
+			pr_info("cpu %d has no node %d or node-local memory\n",
 				cpu, node);
-			if (ptr)
-				printk(KERN_DEBUG "per cpu data for cpu%d at %016lx\n",
-					 cpu, __pa(ptr));
-		}
-		else {
+			pr_debug("per cpu data for cpu%d at %016lx\n",
+				 cpu, __pa(ptr));
+		} else {
 			ptr = __alloc_bootmem_node(NODE_DATA(node), size, align,
 							__pa(MAX_DMA_ADDRESS));
-			if (ptr)
-				printk(KERN_DEBUG "per cpu data for cpu%d on node%d at %016lx\n",
-					 cpu, node, __pa(ptr));
+			pr_debug("per cpu data for cpu%d on node%d at %016lx\n",
+				cpu, node, __pa(ptr));
 		}
 #endif
 		per_cpu_offset(cpu) = ptr - __per_cpu_start;
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 	}
 
-	printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d, nr_node_ids %d\n",
-		NR_CPUS, nr_cpu_ids, nr_node_ids);
-
 	/* Setup percpu data maps */
 	setup_per_cpu_maps();
 
@@ -282,7 +278,7 @@ static void __cpuinit numa_set_cpumask(int cpu, int enable)
 	else
 		cpu_clear(cpu, *mask);
 
-	cpulist_scnprintf(buf, sizeof(buf), *mask);
+	cpulist_scnprintf(buf, sizeof(buf), mask);
 	printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n",
 		enable? "numa_add_cpu":"numa_remove_cpu", cpu, node, buf);
  }
@@ -334,25 +330,25 @@ static const cpumask_t cpu_mask_none;
 /*
  * Returns a pointer to the bitmask of CPUs on Node 'node'.
  */
-const cpumask_t *_node_to_cpumask_ptr(int node)
+const cpumask_t *cpumask_of_node(int node)
 {
 	if (node_to_cpumask_map == NULL) {
 		printk(KERN_WARNING
-			"_node_to_cpumask_ptr(%d): no node_to_cpumask_map!\n",
+			"cpumask_of_node(%d): no node_to_cpumask_map!\n",
 			node);
 		dump_stack();
 		return (const cpumask_t *)&cpu_online_map;
 	}
 	if (node >= nr_node_ids) {
 		printk(KERN_WARNING
-			"_node_to_cpumask_ptr(%d): node > nr_node_ids(%d)\n",
+			"cpumask_of_node(%d): node > nr_node_ids(%d)\n",
 			node, nr_node_ids);
 		dump_stack();
 		return &cpu_mask_none;
 	}
 	return &node_to_cpumask_map[node];
 }
-EXPORT_SYMBOL(_node_to_cpumask_ptr);
+EXPORT_SYMBOL(cpumask_of_node);
 
 /*
  * Returns a bitmask of CPUs on Node 'node'.
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 7e558db362c1..beea2649a240 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -118,22 +118,22 @@ static void native_smp_send_reschedule(int cpu)
 		WARN_ON(1);
 		return;
 	}
-	send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
+	send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR);
 }
 
 void native_send_call_func_single_ipi(int cpu)
 {
-	send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNCTION_SINGLE_VECTOR);
+	send_IPI_mask(cpumask_of(cpu), CALL_FUNCTION_SINGLE_VECTOR);
 }
 
-void native_send_call_func_ipi(cpumask_t mask)
+void native_send_call_func_ipi(const struct cpumask *mask)
 {
 	cpumask_t allbutself;
 
 	allbutself = cpu_online_map;
 	cpu_clear(smp_processor_id(), allbutself);
 
-	if (cpus_equal(mask, allbutself) &&
+	if (cpus_equal(*mask, allbutself) &&
 	    cpus_equal(cpu_online_map, cpu_callout_map))
 		send_IPI_allbutself(CALL_FUNCTION_VECTOR);
 	else
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index f8500c969442..6bd4d9b73870 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -102,14 +102,8 @@ EXPORT_SYMBOL(smp_num_siblings);
 /* Last level cache ID of each logical CPU */
 DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
 
-/* bitmap of online cpus */
-cpumask_t cpu_online_map __read_mostly;
-EXPORT_SYMBOL(cpu_online_map);
-
 cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
 
 /* representing HT siblings of each logical CPU */
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map);
@@ -502,7 +496,7 @@ void __cpuinit set_cpu_sibling_map(int cpu)
 }
 
 /* maps the cpu to the sched domain representing multi-core */
-cpumask_t cpu_coregroup_map(int cpu)
+const struct cpumask *cpu_coregroup_mask(int cpu)
 {
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 	/*
@@ -510,9 +504,14 @@ cpumask_t cpu_coregroup_map(int cpu)
 	 * And for power savings, we return cpu_core_map
 	 */
 	if (sched_mc_power_savings || sched_smt_power_savings)
-		return per_cpu(cpu_core_map, cpu);
+		return &per_cpu(cpu_core_map, cpu);
 	else
-		return c->llc_shared_map;
+		return &c->llc_shared_map;
+}
+
+cpumask_t cpu_coregroup_map(int cpu)
+{
+	return *cpu_coregroup_mask(cpu);
 }
 
 static void impress_friends(void)
@@ -1155,7 +1154,7 @@ static void __init smp_cpu_index_default(void)
 	for_each_possible_cpu(i) {
 		c = &cpu_data(i);
 		/* mark all to hotplug */
-		c->cpu_index = NR_CPUS;
+		c->cpu_index = nr_cpu_ids;
 	}
 }
 
@@ -1260,6 +1259,15 @@ void __init native_smp_cpus_done(unsigned int max_cpus)
 	check_nmi_watchdog();
 }
 
+static int __initdata setup_possible_cpus = -1;
+static int __init _setup_possible_cpus(char *str)
+{
+	get_option(&str, &setup_possible_cpus);
+	return 0;
+}
+early_param("possible_cpus", _setup_possible_cpus);
+
+
 /*
  * cpu_possible_map should be static, it cannot change as cpu's
  * are onlined, or offlined. The reason is per-cpu data-structures
@@ -1272,7 +1280,7 @@ void __init native_smp_cpus_done(unsigned int max_cpus)
  *
  * Three ways to find out the number of additional hotplug CPUs:
  * - If the BIOS specified disabled CPUs in ACPI/mptables use that.
- * - The user can overwrite it with additional_cpus=NUM
+ * - The user can overwrite it with possible_cpus=NUM
  * - Otherwise don't reserve additional CPUs.
  * We do this because additional CPUs waste a lot of memory.
  * -AK
@@ -1285,9 +1293,19 @@ __init void prefill_possible_map(void)
 	if (!num_processors)
 		num_processors = 1;
 
-	possible = num_processors + disabled_cpus;
-	if (possible > NR_CPUS)
-		possible = NR_CPUS;
+	if (setup_possible_cpus == -1)
+		possible = num_processors + disabled_cpus;
+	else
+		possible = setup_possible_cpus;
+
+	total_cpus = max_t(int, possible, num_processors + disabled_cpus);
+
+	if (possible > CONFIG_NR_CPUS) {
+		printk(KERN_WARNING
+			"%d Processors exceeds NR_CPUS limit of %d\n",
+			possible, CONFIG_NR_CPUS);
+		possible = CONFIG_NR_CPUS;
+	}
 
 	printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n",
 		possible, max_t(int, possible - num_processors, 0));
@@ -1352,7 +1370,7 @@ void cpu_disable_common(void)
 	lock_vector_lock();
 	remove_cpu_from_maps(cpu);
 	unlock_vector_lock();
-	fixup_irqs(cpu_online_map);
+	fixup_irqs();
 }
 
 int native_cpu_disable(void)
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index 8da059f949be..ce5054642247 100644
--- a/arch/x86/kernel/tlb_32.c
+++ b/arch/x86/kernel/tlb_32.c
@@ -163,7 +163,7 @@ void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
 	 * We have to send the IPI only to
 	 * CPUs affected.
 	 */
-	send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR);
+	send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR);
 
 	while (!cpus_empty(flush_cpumask))
 		/* nothing. lockup detection does not belong here */
diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c
index 29887d7081a9..f8be6f1d2e48 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -191,7 +191,7 @@ void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
 	 * We have to send the IPI only to
 	 * CPUs affected.
 	 */
-	send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR_START + sender);
+	send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR_START + sender);
 
 	while (!cpus_empty(f->flush_cpumask))
 		cpu_relax();
diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c
index 6a00e5faaa74..f885023167e0 100644
--- a/arch/x86/kernel/tlb_uv.c
+++ b/arch/x86/kernel/tlb_uv.c
@@ -582,7 +582,6 @@ static int __init uv_ptc_init(void)
 static struct bau_control * __init uv_table_bases_init(int blade, int node)
 {
 	int i;
-	int *ip;
 	struct bau_msg_status *msp;
 	struct bau_control *bau_tabp;
 
@@ -599,13 +598,6 @@ static struct bau_control * __init uv_table_bases_init(int blade, int node)
 		bau_cpubits_clear(&msp->seen_by, (int)
 				  uv_blade_nr_possible_cpus(blade));
 
-	bau_tabp->watching =
-	    kmalloc_node(sizeof(int) * DEST_NUM_RESOURCES, GFP_KERNEL, node);
-	BUG_ON(!bau_tabp->watching);
-
-	for (i = 0, ip = bau_tabp->watching; i < DEST_Q_SIZE; i++, ip++)
-		*ip = 0;
-
 	uv_bau_table_bases[blade] = bau_tabp;
 
 	return bau_tabp;
@@ -628,7 +620,6 @@ uv_table_bases_finish(int blade, int node, int cur_cpu,
 		bcp->bau_msg_head	= bau_tablesp->va_queue_first;
 		bcp->va_queue_first	= bau_tablesp->va_queue_first;
 		bcp->va_queue_last	= bau_tablesp->va_queue_last;
-		bcp->watching		= bau_tablesp->watching;
 		bcp->msg_statuses	= bau_tablesp->msg_statuses;
 		bcp->descriptor_base	= adp;
 	}
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 141907ab6e22..c9a666cdd3db 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -20,7 +20,6 @@
 #include <linux/module.h>
 #include <linux/ptrace.h>
 #include <linux/string.h>
-#include <linux/unwind.h>
 #include <linux/delay.h>
 #include <linux/errno.h>
 #include <linux/kexec.h>
@@ -51,7 +50,6 @@
 #include <asm/debugreg.h>
 #include <asm/atomic.h>
 #include <asm/system.h>
-#include <asm/unwind.h>
 #include <asm/traps.h>
 #include <asm/desc.h>
 #include <asm/i387.h>
@@ -72,9 +70,6 @@
 
 #include "cpu/mcheck/mce.h"
 
-DECLARE_BITMAP(used_vectors, NR_VECTORS);
-EXPORT_SYMBOL_GPL(used_vectors);
-
 asmlinkage int system_call(void);
 
 /* Do we ignore FPU interrupts ? */
@@ -89,6 +84,9 @@ gate_desc idt_table[256]
 	__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
 #endif
 
+DECLARE_BITMAP(used_vectors, NR_VECTORS);
+EXPORT_SYMBOL_GPL(used_vectors);
+
 static int ignore_nmis;
 
 static inline void conditional_sti(struct pt_regs *regs)
@@ -292,8 +290,10 @@ dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
 	tsk->thread.error_code = error_code;
 	tsk->thread.trap_no = 8;
 
-	/* This is always a kernel trap and never fixable (and thus must
-	   never return). */
+	/*
+	 * This is always a kernel trap and never fixable (and thus must
+	 * never return).
+	 */
 	for (;;)
 		die(str, regs, error_code);
 }
@@ -520,9 +520,11 @@ dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code)
 }
 
 #ifdef CONFIG_X86_64
-/* Help handler running on IST stack to switch back to user stack
-   for scheduling or signal handling. The actual stack switch is done in
-   entry.S */
+/*
+ * Help handler running on IST stack to switch back to user stack
+ * for scheduling or signal handling. The actual stack switch is done in
+ * entry.S
+ */
 asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 {
 	struct pt_regs *regs = eregs;
@@ -532,8 +534,10 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 	/* Exception from user space */
 	else if (user_mode(eregs))
 		regs = task_pt_regs(current);
-	/* Exception from kernel and interrupts are enabled. Move to
-	   kernel process stack. */
+	/*
+	 * Exception from kernel and interrupts are enabled. Move to
+	 * kernel process stack.
+	 */
 	else if (eregs->flags & X86_EFLAGS_IF)
 		regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs));
 	if (eregs != regs)
@@ -685,12 +689,7 @@ void math_error(void __user *ip)
 	cwd = get_fpu_cwd(task);
 	swd = get_fpu_swd(task);
 
-	err = swd & ~cwd & 0x3f;
-
-#ifdef CONFIG_X86_32
-	if (!err)
-		return;
-#endif
+	err = swd & ~cwd;
 
 	if (err & 0x001) {	/* Invalid op */
 		/*
@@ -708,7 +707,11 @@ void math_error(void __user *ip)
 	} else if (err & 0x020) { /* Precision */
 		info.si_code = FPE_FLTRES;
 	} else {
-		info.si_code = __SI_FAULT|SI_KERNEL; /* WTF? */
+		/*
+		 * If we're using IRQ 13, or supposedly even some trap 16
+		 * implementations, it's possible we get a spurious trap...
+		 */
+		return;		/* Spurious trap, no error */
 	}
 	force_sig_info(SIGFPE, &info, task);
 }
@@ -941,9 +944,7 @@ dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code)
 
 void __init trap_init(void)
 {
-#ifdef CONFIG_X86_32
 	int i;
-#endif
 
 #ifdef CONFIG_EISA
 	void __iomem *p = early_ioremap(0x0FFFD9, 4);
@@ -1000,11 +1001,15 @@ void __init trap_init(void)
 	}
 
 	set_system_trap_gate(SYSCALL_VECTOR, &system_call);
+#endif
 
 	/* Reserve all the builtin and the syscall vector: */
 	for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++)
 		set_bit(i, used_vectors);
 
+#ifdef CONFIG_X86_64
+	set_bit(IA32_SYSCALL_VECTOR, used_vectors);
+#else
 	set_bit(SYSCALL_VECTOR, used_vectors);
 #endif
 	/*
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
index 254ee07f8635..c4c1f9e09402 100644
--- a/arch/x86/kernel/vmiclock_32.c
+++ b/arch/x86/kernel/vmiclock_32.c
@@ -226,7 +226,7 @@ static void __devinit vmi_time_init_clockevent(void)
 	/* Upper bound is clockevent's use of ulong for cycle deltas. */
 	evt->max_delta_ns = clockevent_delta2ns(ULONG_MAX, evt);
 	evt->min_delta_ns = clockevent_delta2ns(1, evt);
-	evt->cpumask = cpumask_of_cpu(cpu);
+	evt->cpumask = cpumask_of(cpu);
 
 	printk(KERN_WARNING "vmi: registering clock event %s. mult=%lu shift=%u\n",
 	       evt->name, evt->mult, evt->shift);
diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c
index 15c3e6999182..2b54fe002e94 100644
--- a/arch/x86/kernel/xsave.c
+++ b/arch/x86/kernel/xsave.c
@@ -159,7 +159,7 @@ int save_i387_xstate(void __user *buf)
  * Restore the extended state if present. Otherwise, restore the FP/SSE
  * state.
  */
-int restore_user_xstate(void __user *buf)
+static int restore_user_xstate(void __user *buf)
 {
 	struct _fpx_sw_bytes fx_sw_user;
 	u64 mask;