Merge commit 'rr_cpumask/master'

Conflicts:
	kernel/sched.c

35 files changed, 536 insertions, 367 deletions

diff --git a/Documentation/cputopology.txt b/Documentation/cputopology.txt
index bd699da24666..45932ec21cee 100644
--- a/Documentation/cputopology.txt
+++ b/Documentation/cputopology.txt
@@ -31,3 +31,51 @@ not defined by include/asm-XXX/topology.h:
 2) core_id: 0
 3) thread_siblings: just the given CPU
 4) core_siblings: just the given CPU
+
+Additionally, cpu topology information is provided under
+/sys/devices/system/cpu and includes these files.  The internal
+source for the output is in brackets ("[]").
+
+    kernel_max: the maximum cpu index allowed by the kernel configuration.
+		[NR_CPUS-1]
+
+    offline:	cpus that are not online because they have been
+		HOTPLUGGED off (see cpu-hotplug.txt) or exceed the limit
+		of cpus allowed by the kernel configuration (kernel_max
+		above). [~cpu_online_mask + cpus >= NR_CPUS]
+
+    online:	cpus that are online and being scheduled [cpu_online_mask]
+
+    possible:	cpus that have been allocated resources and can be
+		brought online if they are present. [cpu_possible_mask]
+
+    present:	cpus that have been identified as being present in the
+		system. [cpu_present_mask]
+
+The format for the above output is compatible with cpulist_parse()
+[see <linux/cpumask.h>].  Some examples follow.
+
+In this example, there are 64 cpus in the system but cpus 32-63 exceed
+the kernel max which is limited to 0..31 by the NR_CPUS config option
+being 32.  Note also that cpus 2 and 4-31 are not online but could be
+brought online as they are both present and possible.
+
+     kernel_max: 31
+        offline: 2,4-31,32-63
+         online: 0-1,3
+       possible: 0-31
+        present: 0-31
+
+In this example, the NR_CPUS config option is 128, but the kernel was
+started with possible_cpus=144.  There are 4 cpus in the system and cpu2
+was manually taken offline (and is the only cpu that can be brought
+online.)
+
+     kernel_max: 127
+        offline: 2,4-127,128-143
+         online: 0-1,3
+       possible: 0-127
+        present: 0-3
+
+See cpu-hotplug.txt for the possible_cpus=NUM kernel start parameter
+as well as more information on the various cpumask's.
diff --git a/arch/alpha/include/asm/topology.h b/arch/alpha/include/asm/topology.h
index 149532e162c4..b4f284c72ff3 100644
--- a/arch/alpha/include/asm/topology.h
+++ b/arch/alpha/include/asm/topology.h
@@ -39,7 +39,24 @@ static inline cpumask_t node_to_cpumask(int node)
 	return node_cpu_mask;
 }
 
+extern struct cpumask node_to_cpumask_map[];
+/* FIXME: This is dumb, recalculating every time.  But simple. */
+static const struct cpumask *cpumask_of_node(int node)
+{
+	int cpu;
+
+	cpumask_clear(&node_to_cpumask_map[node]);
+
+	for_each_online_cpu(cpu) {
+		if (cpu_to_node(cpu) == node)
+			cpumask_set_cpu(cpu, node_to_cpumask_map[node]);
+	}
+
+	return &node_to_cpumask_map[node];
+}
+
 #define pcibus_to_cpumask(bus)	(cpu_online_map)
+#define cpumask_of_pcibus(bus)	(cpu_online_mask)
 
 #endif /* !CONFIG_NUMA */
 # include <asm-generic/topology.h>
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index a449e999027c..02bee6983ce2 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -79,6 +79,11 @@ int alpha_l3_cacheshape;
 unsigned long alpha_verbose_mcheck = CONFIG_VERBOSE_MCHECK_ON;
 #endif
 
+#ifdef CONFIG_NUMA
+struct cpumask node_to_cpumask_map[MAX_NUMNODES] __read_mostly;
+EXPORT_SYMBOL(node_to_cpumask_map);
+#endif
+
 /* Which processor we booted from.  */
 int boot_cpuid;
 
diff --git a/arch/ia64/include/asm/topology.h b/arch/ia64/include/asm/topology.h
index a3cc9f65f954..2375e524a186 100644
--- a/arch/ia64/include/asm/topology.h
+++ b/arch/ia64/include/asm/topology.h
@@ -34,6 +34,7 @@
  * Returns a bitmask of CPUs on Node 'node'.
  */
 #define node_to_cpumask(node) (node_to_cpu_mask[node])
+#define cpumask_of_node(node) (&node_to_cpu_mask[node])
 
 /*
  * Returns the number of the node containing Node 'nid'.
@@ -45,7 +46,7 @@
 /*
  * Returns the number of the first CPU on Node 'node'.
  */
-#define node_to_first_cpu(node) (first_cpu(node_to_cpumask(node)))
+#define node_to_first_cpu(node) (cpumask_first(cpumask_of_node(node)))
 
 /*
  * Determines the node for a given pci bus
@@ -119,6 +120,10 @@ extern void arch_fix_phys_package_id(int num, u32 slot);
 					node_to_cpumask(pcibus_to_node(bus)) \
 				)
 
+#define cpumask_of_pcibus(bus)	(pcibus_to_node(bus) == -1 ?		\
+				 cpu_all_mask :				\
+				 cpumask_from_node(pcibus_to_node(bus)))
+
 #include <asm-generic/topology.h>
 
 #endif /* _ASM_IA64_TOPOLOGY_H */
diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index bd7acc71e8a9..54ae373e6e22 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -1001,7 +1001,7 @@ acpi_map_iosapic(acpi_handle handle, u32 depth, void *context, void **ret)
 	node = pxm_to_node(pxm);
 
 	if (node >= MAX_NUMNODES || !node_online(node) ||
-	    cpus_empty(node_to_cpumask(node)))
+	    cpumask_empty(cpumask_of_node(node)))
 		return AE_OK;
 
 	/* We know a gsi to node mapping! */
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index c8adecd5b416..5cfd3d91001a 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -695,32 +695,31 @@ get_target_cpu (unsigned int gsi, int irq)
 #ifdef CONFIG_NUMA
 	{
 		int num_cpus, cpu_index, iosapic_index, numa_cpu, i = 0;
-		cpumask_t cpu_mask;
+		const struct cpumask *cpu_mask;
 
 		iosapic_index = find_iosapic(gsi);
 		if (iosapic_index < 0 ||
 		    iosapic_lists[iosapic_index].node == MAX_NUMNODES)
 			goto skip_numa_setup;
 
-		cpu_mask = node_to_cpumask(iosapic_lists[iosapic_index].node);
-		cpus_and(cpu_mask, cpu_mask, domain);
-		for_each_cpu_mask(numa_cpu, cpu_mask) {
-			if (!cpu_online(numa_cpu))
-				cpu_clear(numa_cpu, cpu_mask);
+		cpu_mask = cpumask_of_node(iosapic_lists[iosapic_index].node);
+		num_cpus = 0;
+		for_each_cpu_and(numa_cpu, cpu_mask, &domain) {
+			if (cpu_online(numa_cpu))
+				num_cpus++;
 		}
 
-		num_cpus = cpus_weight(cpu_mask);
-
 		if (!num_cpus)
 			goto skip_numa_setup;
 
 		/* Use irq assignment to distribute across cpus in node */
 		cpu_index = irq % num_cpus;
 
-		for (numa_cpu = first_cpu(cpu_mask) ; i < cpu_index ; i++)
-			numa_cpu = next_cpu(numa_cpu, cpu_mask);
+		for_each_cpu_and(numa_cpu, cpu_mask, &domain)
+			if (cpu_online(numa_cpu) && i++ >= cpu_index)
+				break;
 
-		if (numa_cpu != NR_CPUS)
+		if (numa_cpu < nr_cpu_ids)
 			return cpu_physical_id(numa_cpu);
 	}
 skip_numa_setup:
@@ -731,7 +730,7 @@ skip_numa_setup:
 	 * case of NUMA.)
 	 */
 	do {
-		if (++cpu >= NR_CPUS)
+		if (++cpu >= nr_cpu_ids)
 			cpu = 0;
 	} while (!cpu_online(cpu) || !cpu_isset(cpu, domain));
 
diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
index 636588e7e068..be339477f906 100644
--- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c
+++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
@@ -385,7 +385,6 @@ static int sn_topology_show(struct seq_file *s, void *d)
 	int j;
 	const char *slabname;
 	int ordinal;
-	cpumask_t cpumask;
 	char slice;
 	struct cpuinfo_ia64 *c;
 	struct sn_hwperf_port_info *ptdata;
@@ -473,23 +472,21 @@ static int sn_topology_show(struct seq_file *s, void *d)
 		 * CPUs on this node, if any
 		 */
 		if (!SN_HWPERF_IS_IONODE(obj)) {
-			cpumask = node_to_cpumask(ordinal);
-			for_each_online_cpu(i) {
-				if (cpu_isset(i, cpumask)) {
-					slice = 'a' + cpuid_to_slice(i);
-					c = cpu_data(i);
-					seq_printf(s, "cpu %d %s%c local"
-						" freq %luMHz, arch ia64",
-						i, obj->location, slice,
-						c->proc_freq / 1000000);
-					for_each_online_cpu(j) {
-						seq_printf(s, j ? ":%d" : ", dist %d",
-							node_distance(
+			for_each_cpu_and(i, cpu_online_mask,
+					 cpumask_of_node(ordinal)) {
+				slice = 'a' + cpuid_to_slice(i);
+				c = cpu_data(i);
+				seq_printf(s, "cpu %d %s%c local"
+					   " freq %luMHz, arch ia64",
+					   i, obj->location, slice,
+					   c->proc_freq / 1000000);
+				for_each_online_cpu(j) {
+					seq_printf(s, j ? ":%d" : ", dist %d",
+						   node_distance(
 						    	cpu_to_node(i),
 						    	cpu_to_node(j)));
-					}
-					seq_putc(s, '\n');
 				}
+				seq_putc(s, '\n');
 			}
 		}
 	}
diff --git a/arch/mips/include/asm/mach-ip27/topology.h b/arch/mips/include/asm/mach-ip27/topology.h
index 1fb959f98982..55d481569a1f 100644
--- a/arch/mips/include/asm/mach-ip27/topology.h
+++ b/arch/mips/include/asm/mach-ip27/topology.h
@@ -25,11 +25,13 @@ extern struct cpuinfo_ip27 sn_cpu_info[NR_CPUS];
 #define cpu_to_node(cpu)	(sn_cpu_info[(cpu)].p_nodeid)
 #define parent_node(node)	(node)
 #define node_to_cpumask(node)	(hub_data(node)->h_cpus)
-#define node_to_first_cpu(node)	(first_cpu(node_to_cpumask(node)))
+#define cpumask_of_node(node)	(&hub_data(node)->h_cpus)
+#define node_to_first_cpu(node)	(cpumask_first(cpumask_of_node(node)))
 struct pci_bus;
 extern int pcibus_to_node(struct pci_bus *);
 
 #define pcibus_to_cpumask(bus)	(cpu_online_map)
+#define cpumask_of_pcibus(bus)	(cpu_online_mask)
 
 extern unsigned char __node_distances[MAX_COMPACT_NODES][MAX_COMPACT_NODES];
 
diff --git a/arch/powerpc/include/asm/topology.h b/arch/powerpc/include/asm/topology.h
index 373fca394a54..8bb79a673c5c 100644
--- a/arch/powerpc/include/asm/topology.h
+++ b/arch/powerpc/include/asm/topology.h
@@ -22,11 +22,11 @@ static inline cpumask_t node_to_cpumask(int node)
 	return numa_cpumask_lookup_table[node];
 }
 
+#define cpumask_of_node(node) (&numa_cpumask_lookup_table[node])
+
 static inline int node_to_first_cpu(int node)
 {
-	cpumask_t tmp;
-	tmp = node_to_cpumask(node);
-	return first_cpu(tmp);
+	return cpumask_first(cpumask_of_node(node));
 }
 
 int of_node_to_nid(struct device_node *device);
@@ -46,6 +46,10 @@ static inline int pcibus_to_node(struct pci_bus *bus)
 					node_to_cpumask(pcibus_to_node(bus)) \
 				)
 
+#define cpumask_of_pcibus(bus)	(pcibus_to_node(bus) == -1 ?		\
+				 cpu_all_mask :				\
+				 cpumask_of_node(pcibus_to_node(bus)))
+
 /* sched_domains SD_NODE_INIT for PPC64 machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
 	.parent			= NULL,			\
diff --git a/arch/powerpc/platforms/cell/spu_priv1_mmio.c b/arch/powerpc/platforms/cell/spu_priv1_mmio.c
index 906a0a2a9fe1..1410443731eb 100644
--- a/arch/powerpc/platforms/cell/spu_priv1_mmio.c
+++ b/arch/powerpc/platforms/cell/spu_priv1_mmio.c
@@ -80,10 +80,10 @@ static void cpu_affinity_set(struct spu *spu, int cpu)
 	u64 route;
 
 	if (nr_cpus_node(spu->node)) {
-		cpumask_t spumask = node_to_cpumask(spu->node);
-		cpumask_t cpumask = node_to_cpumask(cpu_to_node(cpu));
+		const struct cpumask *spumask = cpumask_of_node(spu->node),
+			*cpumask = cpumask_of_node(cpu_to_node(cpu));
 
-		if (!cpus_intersects(spumask, cpumask))
+		if (!cpumask_intersects(spumask, cpumask))
 			return;
 	}
 
diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c
index 2ad914c47493..6a0ad196aeb3 100644
--- a/arch/powerpc/platforms/cell/spufs/sched.c
+++ b/arch/powerpc/platforms/cell/spufs/sched.c
@@ -166,9 +166,9 @@ void spu_update_sched_info(struct spu_context *ctx)
 static int __node_allowed(struct spu_context *ctx, int node)
 {
 	if (nr_cpus_node(node)) {
-		cpumask_t mask = node_to_cpumask(node);
+		const struct cpumask *mask = cpumask_of_node(node);
 
-		if (cpus_intersects(mask, ctx->cpus_allowed))
+		if (cpumask_intersects(mask, &ctx->cpus_allowed))
 			return 1;
 	}
 
diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h
index d96c91643458..fff4a86c602a 100644
--- a/arch/s390/include/asm/topology.h
+++ b/arch/s390/include/asm/topology.h
@@ -6,6 +6,7 @@
 #define mc_capable()	(1)
 
 cpumask_t cpu_coregroup_map(unsigned int cpu);
+const struct cpumask *cpu_coregroup_mask(unsigned int cpu);
 
 extern cpumask_t cpu_core_map[NR_CPUS];
 
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 90e9ba11eba1..cc362c9ea8f1 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -97,6 +97,11 @@ cpumask_t cpu_coregroup_map(unsigned int cpu)
 	return mask;
 }
 
+const struct cpumask *cpu_coregroup_mask(unsigned int cpu)
+{
+	return &cpu_core_map[cpu];
+}
+
 static void add_cpus_to_core(struct tl_cpu *tl_cpu, struct core_info *core)
 {
 	unsigned int cpu;
diff --git a/arch/sh/include/asm/topology.h b/arch/sh/include/asm/topology.h
index 279d9cc4a007..066f0fba590e 100644
--- a/arch/sh/include/asm/topology.h
+++ b/arch/sh/include/asm/topology.h
@@ -32,6 +32,7 @@
 #define parent_node(node)	((void)(node),0)
 
 #define node_to_cpumask(node)	((void)node, cpu_online_map)
+#define cpumask_of_node(node)	((void)node, cpu_online_mask)
 #define node_to_first_cpu(node)	((void)(node),0)
 
 #define pcibus_to_node(bus)	((void)(bus), -1)
diff --git a/arch/sparc/include/asm/topology_64.h b/arch/sparc/include/asm/topology_64.h
index 001c04027c82..3270cfb1ab53 100644
--- a/arch/sparc/include/asm/topology_64.h
+++ b/arch/sparc/include/asm/topology_64.h
@@ -16,8 +16,12 @@ static inline cpumask_t node_to_cpumask(int node)
 {
 	return numa_cpumask_lookup_table[node];
 }
+#define cpumask_of_node(node) (&numa_cpumask_lookup_table[node])
 
-/* Returns a pointer to the cpumask of CPUs on Node 'node'. */
+/*
+ * Returns a pointer to the cpumask of CPUs on Node 'node'.
+ * Deprecated: use "const struct cpumask *mask = cpumask_of_node(node)"
+ */
 #define node_to_cpumask_ptr(v, node)		\
 		cpumask_t *v = &(numa_cpumask_lookup_table[node])
 
@@ -26,9 +30,7 @@ static inline cpumask_t node_to_cpumask(int node)
 
 static inline int node_to_first_cpu(int node)
 {
-	cpumask_t tmp;
-	tmp = node_to_cpumask(node);
-	return first_cpu(tmp);
+	return cpumask_first(cpumask_of_node(node));
 }
 
 struct pci_bus;
@@ -82,5 +84,6 @@ static inline int pcibus_to_node(struct pci_bus *pbus)
 #endif /* CONFIG_SMP */
 
 #define cpu_coregroup_map(cpu)			(cpu_core_map[cpu])
+#define cpu_coregroup_mask(cpu)			(&cpu_core_map[cpu])
 
 #endif /* _ASM_SPARC64_TOPOLOGY_H */
diff --git a/arch/sparc/kernel/of_device_64.c b/arch/sparc/kernel/of_device_64.c
index 322046cdf85f..4873f28905b0 100644
--- a/arch/sparc/kernel/of_device_64.c
+++ b/arch/sparc/kernel/of_device_64.c
@@ -778,7 +778,7 @@ static unsigned int __init build_one_device_irq(struct of_device *op,
 out:
 	nid = of_node_to_nid(dp);
 	if (nid != -1) {
-		cpumask_t numa_mask = node_to_cpumask(nid);
+		cpumask_t numa_mask = *cpumask_of_node(nid);
 
 		irq_set_affinity(irq, &numa_mask);
 	}
diff --git a/arch/sparc/kernel/pci_msi.c b/arch/sparc/kernel/pci_msi.c
index 0d0cd815e83e..4ef282e81912 100644
--- a/arch/sparc/kernel/pci_msi.c
+++ b/arch/sparc/kernel/pci_msi.c
@@ -286,7 +286,7 @@ static int bringup_one_msi_queue(struct pci_pbm_info *pbm,
 
 	nid = pbm->numa_node;
 	if (nid != -1) {
-		cpumask_t numa_mask = node_to_cpumask(nid);
+		cpumask_t numa_mask = *cpumask_of_node(nid);
 
 		irq_set_affinity(irq, &numa_mask);
 	}
diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h
index 647781298e7e..f8959c7a985f 100644
--- a/arch/x86/include/asm/pci.h
+++ b/arch/x86/include/asm/pci.h
@@ -100,9 +100,9 @@ static inline void early_quirks(void) { }
 
 #ifdef CONFIG_NUMA
 /* Returns the node based on pci bus */
-static inline int __pcibus_to_node(struct pci_bus *bus)
+static inline int __pcibus_to_node(const struct pci_bus *bus)
 {
-	struct pci_sysdata *sd = bus->sysdata;
+	const struct pci_sysdata *sd = bus->sysdata;
 
 	return sd->node;
 }
@@ -111,6 +111,12 @@ static inline cpumask_t __pcibus_to_cpumask(struct pci_bus *bus)
 {
 	return node_to_cpumask(__pcibus_to_node(bus));
 }
+
+static inline const struct cpumask *
+cpumask_of_pcibus(const struct pci_bus *bus)
+{
+	return cpumask_of_node(__pcibus_to_node(bus));
+}
 #endif
 
 #endif /* _ASM_X86_PCI_H */
diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h
index 79e31e9dcdda..4e2f2e0aab27 100644
--- a/arch/x86/include/asm/topology.h
+++ b/arch/x86/include/asm/topology.h
@@ -61,13 +61,19 @@ static inline int cpu_to_node(int cpu)
  *
  * Side note: this function creates the returned cpumask on the stack
  * so with a high NR_CPUS count, excessive stack space is used.  The
- * node_to_cpumask_ptr function should be used whenever possible.
+ * cpumask_of_node function should be used whenever possible.
  */
 static inline cpumask_t node_to_cpumask(int node)
 {
 	return node_to_cpumask_map[node];
 }
 
+/* Returns a bitmask of CPUs on Node 'node'. */
+static inline const struct cpumask *cpumask_of_node(int node)
+{
+	return &node_to_cpumask_map[node];
+}
+
 #else /* CONFIG_X86_64 */
 
 /* Mappings between node number and cpus on that node. */
@@ -82,7 +88,7 @@ DECLARE_EARLY_PER_CPU(int, x86_cpu_to_node_map);
 #ifdef CONFIG_DEBUG_PER_CPU_MAPS
 extern int cpu_to_node(int cpu);
 extern int early_cpu_to_node(int cpu);
-extern const cpumask_t *_node_to_cpumask_ptr(int node);
+extern const cpumask_t *cpumask_of_node(int node);
 extern cpumask_t node_to_cpumask(int node);
 
 #else	/* !CONFIG_DEBUG_PER_CPU_MAPS */
@@ -103,7 +109,7 @@ static inline int early_cpu_to_node(int cpu)
 }
 
 /* Returns a pointer to the cpumask of CPUs on Node 'node'. */
-static inline const cpumask_t *_node_to_cpumask_ptr(int node)
+static inline const cpumask_t *cpumask_of_node(int node)
 {
 	return &node_to_cpumask_map[node];
 }
@@ -116,12 +122,15 @@ static inline cpumask_t node_to_cpumask(int node)
 
 #endif /* !CONFIG_DEBUG_PER_CPU_MAPS */
 
-/* Replace default node_to_cpumask_ptr with optimized version */
+/*
+ * Replace default node_to_cpumask_ptr with optimized version
+ * Deprecated: use "const struct cpumask *mask = cpumask_of_node(node)"
+ */
 #define node_to_cpumask_ptr(v, node)		\
-		const cpumask_t *v = _node_to_cpumask_ptr(node)
+		const cpumask_t *v = cpumask_of_node(node)
 
 #define node_to_cpumask_ptr_next(v, node)	\
-			   v = _node_to_cpumask_ptr(node)
+			   v = cpumask_of_node(node)
 
 #endif /* CONFIG_X86_64 */
 
@@ -187,7 +196,7 @@ extern int __node_distance(int, int);
 #define	cpu_to_node(cpu)	0
 #define	early_cpu_to_node(cpu)	0
 
-static inline const cpumask_t *_node_to_cpumask_ptr(int node)
+static inline const cpumask_t *cpumask_of_node(int node)
 {
 	return &cpu_online_map;
 }
@@ -200,12 +209,15 @@ static inline int node_to_first_cpu(int node)
 	return first_cpu(cpu_online_map);
 }
 
-/* Replace default node_to_cpumask_ptr with optimized version */
+/*
+ * Replace default node_to_cpumask_ptr with optimized version
+ * Deprecated: use "const struct cpumask *mask = cpumask_of_node(node)"
+ */
 #define node_to_cpumask_ptr(v, node)		\
-		const cpumask_t *v = _node_to_cpumask_ptr(node)
+		const cpumask_t *v = cpumask_of_node(node)
 
 #define node_to_cpumask_ptr_next(v, node)	\
-			   v = _node_to_cpumask_ptr(node)
+			   v = cpumask_of_node(node)
 #endif
 
 #include <asm-generic/topology.h>
@@ -214,12 +226,12 @@ static inline int node_to_first_cpu(int node)
 /* Returns the number of the first CPU on Node 'node'. */
 static inline int node_to_first_cpu(int node)
 {
-	node_to_cpumask_ptr(mask, node);
-	return first_cpu(*mask);
+	return cpumask_first(cpumask_of_node(node));
 }
 #endif
 
 extern cpumask_t cpu_coregroup_map(int cpu);
+extern const struct cpumask *cpu_coregroup_mask(int cpu);
 
 #ifdef ENABLE_TOPO_DEFINES
 #define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index 0b63b08e7530..49f3f709ee1f 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -339,25 +339,25 @@ 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/smpboot.c b/arch/x86/kernel/smpboot.c
index 31869bf5fabd..9e177a4077ee 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -496,7 +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);
 	/*
@@ -504,9 +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)
diff --git a/block/blk.h b/block/blk.h
index d2e49af90db5..6e1ed40534e9 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -99,8 +99,8 @@ static inline int queue_congestion_off_threshold(struct request_queue *q)
 static inline int blk_cpu_to_group(int cpu)
 {
 #ifdef CONFIG_SCHED_MC
-	cpumask_t mask = cpu_coregroup_map(cpu);
-	return first_cpu(mask);
+	const struct cpumask *mask = cpu_coregroup_mask(cpu);
+	return cpumask_first(mask);
 #elif defined(CONFIG_SCHED_SMT)
 	return first_cpu(per_cpu(cpu_sibling_map, cpu));
 #else
diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 4259072f5bd0..2aef96f20b30 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -128,10 +128,54 @@ print_cpus_func(online);
 print_cpus_func(possible);
 print_cpus_func(present);
 
+/*
+ * Print values for NR_CPUS and offlined cpus
+ */
+static ssize_t print_cpus_kernel_max(struct sysdev_class *class, char *buf)
+{
+	int n = snprintf(buf, PAGE_SIZE-2, "%d\n", CONFIG_NR_CPUS - 1);
+	return n;
+}
+static SYSDEV_CLASS_ATTR(kernel_max, 0444, print_cpus_kernel_max, NULL);
+
+/* arch-optional setting to enable display of offline cpus >= nr_cpu_ids */
+unsigned int total_cpus;
+
+static ssize_t print_cpus_offline(struct sysdev_class *class, char *buf)
+{
+	int n = 0, len = PAGE_SIZE-2;
+	cpumask_var_t offline;
+
+	/* display offline cpus < nr_cpu_ids */
+	if (!alloc_cpumask_var(&offline, GFP_KERNEL))
+		return -ENOMEM;
+	cpumask_complement(offline, cpu_online_mask);
+	n = cpulist_scnprintf(buf, len, offline);
+	free_cpumask_var(offline);
+
+	/* display offline cpus >= nr_cpu_ids */
+	if (total_cpus && nr_cpu_ids < total_cpus) {
+		if (n && n < len)
+			buf[n++] = ',';
+
+		if (nr_cpu_ids == total_cpus-1)
+			n += snprintf(&buf[n], len - n, "%d", nr_cpu_ids);
+		else
+			n += snprintf(&buf[n], len - n, "%d-%d",
+						      nr_cpu_ids, total_cpus-1);
+	}
+
+	n += snprintf(&buf[n], len - n, "\n");
+	return n;
+}
+static SYSDEV_CLASS_ATTR(offline, 0444, print_cpus_offline, NULL);
+
 static struct sysdev_class_attribute *cpu_state_attr[] = {
 	&attr_online_map,
 	&attr_possible_map,
 	&attr_present_map,
+	&attr_kernel_max,
+	&attr_offline,
 };
 
 static int cpu_states_init(void)
diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c
index 757035ea246f..3128a5090dbd 100644
--- a/drivers/infiniband/hw/ehca/ehca_irq.c
+++ b/drivers/infiniband/hw/ehca/ehca_irq.c
@@ -659,12 +659,12 @@ static inline int find_next_online_cpu(struct ehca_comp_pool *pool)
 
 	WARN_ON_ONCE(!in_interrupt());
 	if (ehca_debug_level >= 3)
-		ehca_dmp(&cpu_online_map, sizeof(cpumask_t), "");
+		ehca_dmp(cpu_online_mask, cpumask_size(), "");
 
 	spin_lock_irqsave(&pool->last_cpu_lock, flags);
-	cpu = next_cpu_nr(pool->last_cpu, cpu_online_map);
+	cpu = cpumask_next(pool->last_cpu, cpu_online_mask);
 	if (cpu >= nr_cpu_ids)
-		cpu = first_cpu(cpu_online_map);
+		cpu = cpumask_first(cpu_online_mask);
 	pool->last_cpu = cpu;
 	spin_unlock_irqrestore(&pool->last_cpu_lock, flags);
 
@@ -855,7 +855,7 @@ static int __cpuinit comp_pool_callback(struct notifier_block *nfb,
 	case CPU_UP_CANCELED_FROZEN:
 		ehca_gen_dbg("CPU: %x (CPU_CANCELED)", cpu);
 		cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
-		kthread_bind(cct->task, any_online_cpu(cpu_online_map));
+		kthread_bind(cct->task, cpumask_any(cpu_online_mask));
 		destroy_comp_task(pool, cpu);
 		break;
 	case CPU_ONLINE:
@@ -902,7 +902,7 @@ int ehca_create_comp_pool(void)
 		return -ENOMEM;
 
 	spin_lock_init(&pool->last_cpu_lock);
-	pool->last_cpu = any_online_cpu(cpu_online_map);
+	pool->last_cpu = cpumask_any(cpu_online_mask);
 
 	pool->cpu_comp_tasks = alloc_percpu(struct ehca_cpu_comp_task);
 	if (pool->cpu_comp_tasks == NULL) {
@@ -934,10 +934,9 @@ void ehca_destroy_comp_pool(void)
 
 	unregister_hotcpu_notifier(&comp_pool_callback_nb);
 
-	for (i = 0; i < NR_CPUS; i++) {
-		if (cpu_online(i))
-			destroy_comp_task(pool, i);
-	}
+	for_each_online_cpu(i)
+		destroy_comp_task(pool, i);
+
 	free_percpu(pool->cpu_comp_tasks);
 	kfree(pool);
 }
diff --git a/drivers/infiniband/hw/ipath/ipath_file_ops.c b/drivers/infiniband/hw/ipath/ipath_file_ops.c
index 239d4e8068ac..23173982b32c 100644
--- a/drivers/infiniband/hw/ipath/ipath_file_ops.c
+++ b/drivers/infiniband/hw/ipath/ipath_file_ops.c
@@ -1679,7 +1679,7 @@ static int find_best_unit(struct file *fp,
 	 * InfiniPath chip to that processor (we assume reasonable connectivity,
 	 * for now).  This code assumes that if affinity has been set
 	 * before this point, that at most one cpu is set; for now this
-	 * is reasonable.  I check for both cpus_empty() and cpus_full(),
+	 * is reasonable.  I check for both cpumask_empty() and cpumask_full(),
 	 * in case some kernel variant sets none of the bits when no
 	 * affinity is set.  2.6.11 and 12 kernels have all present
 	 * cpus set.  Some day we'll have to fix it up further to handle
@@ -1688,11 +1688,11 @@ static int find_best_unit(struct file *fp,
 	 * information.  There may be some issues with dual core numbering
 	 * as well.  This needs more work prior to release.
 	 */
-	if (!cpus_empty(current->cpus_allowed) &&
-	    !cpus_full(current->cpus_allowed)) {
+	if (!cpumask_empty(&current->cpus_allowed) &&
+	    !cpumask_full(&current->cpus_allowed)) {
 		int ncpus = num_online_cpus(), curcpu = -1, nset = 0;
 		for (i = 0; i < ncpus; i++)
-			if (cpu_isset(i, current->cpus_allowed)) {
+			if (cpumask_test_cpu(i, &current->cpus_allowed)) {
 				ipath_cdbg(PROC, "%s[%u] affinity set for "
 					   "cpu %d/%d\n", current->comm,
 					   current->pid, i, ncpus);
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 16c211558c22..c99358a52176 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -462,7 +462,8 @@ int seq_dentry(struct seq_file *m, struct dentry *dentry, char *esc)
 	return -1;
 }
 
-int seq_bitmap(struct seq_file *m, unsigned long *bits, unsigned int nr_bits)
+int seq_bitmap(struct seq_file *m, const unsigned long *bits,
+				   unsigned int nr_bits)
 {
 	if (m->count < m->size) {
 		int len = bitmap_scnprintf(m->buf + m->count,
diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h
index a08c33a26ca9..2878811c6134 100644
--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -137,9 +137,12 @@ extern void bitmap_copy_le(void *dst, const unsigned long *src, int nbits);
 		(1UL<<((nbits) % BITS_PER_LONG))-1 : ~0UL		\
 )
 
+#define small_const_nbits(nbits) \
+	(__builtin_constant_p(nbits) && (nbits) <= BITS_PER_LONG)
+
 static inline void bitmap_zero(unsigned long *dst, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = 0UL;
 	else {
 		int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
@@ -150,7 +153,7 @@ static inline void bitmap_zero(unsigned long *dst, int nbits)
 static inline void bitmap_fill(unsigned long *dst, int nbits)
 {
 	size_t nlongs = BITS_TO_LONGS(nbits);
-	if (nlongs > 1) {
+	if (!small_const_nbits(nbits)) {
 		int len = (nlongs - 1) * sizeof(unsigned long);
 		memset(dst, 0xff,  len);
 	}
@@ -160,7 +163,7 @@ static inline void bitmap_fill(unsigned long *dst, int nbits)
 static inline void bitmap_copy(unsigned long *dst, const unsigned long *src,
 			int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src;
 	else {
 		int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
@@ -171,7 +174,7 @@ static inline void bitmap_copy(unsigned long *dst, const unsigned long *src,
 static inline void bitmap_and(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 & *src2;
 	else
 		__bitmap_and(dst, src1, src2, nbits);
@@ -180,7 +183,7 @@ static inline void bitmap_and(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_or(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 | *src2;
 	else
 		__bitmap_or(dst, src1, src2, nbits);
@@ -189,7 +192,7 @@ static inline void bitmap_or(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_xor(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 ^ *src2;
 	else
 		__bitmap_xor(dst, src1, src2, nbits);
@@ -198,7 +201,7 @@ static inline void bitmap_xor(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_andnot(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 & ~(*src2);
 	else
 		__bitmap_andnot(dst, src1, src2, nbits);
@@ -207,7 +210,7 @@ static inline void bitmap_andnot(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_complement(unsigned long *dst, const unsigned long *src,
 			int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = ~(*src) & BITMAP_LAST_WORD_MASK(nbits);
 	else
 		__bitmap_complement(dst, src, nbits);
@@ -216,7 +219,7 @@ static inline void bitmap_complement(unsigned long *dst, const unsigned long *sr
 static inline int bitmap_equal(const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! ((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_equal(src1, src2, nbits);
@@ -225,7 +228,7 @@ static inline int bitmap_equal(const unsigned long *src1,
 static inline int bitmap_intersects(const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ((*src1 & *src2) & BITMAP_LAST_WORD_MASK(nbits)) != 0;
 	else
 		return __bitmap_intersects(src1, src2, nbits);
@@ -234,7 +237,7 @@ static inline int bitmap_intersects(const unsigned long *src1,
 static inline int bitmap_subset(const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! ((*src1 & ~(*src2)) & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_subset(src1, src2, nbits);
@@ -242,7 +245,7 @@ static inline int bitmap_subset(const unsigned long *src1,
 
 static inline int bitmap_empty(const unsigned long *src, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! (*src & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_empty(src, nbits);
@@ -250,7 +253,7 @@ static inline int bitmap_empty(const unsigned long *src, int nbits)
 
 static inline int bitmap_full(const unsigned long *src, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! (~(*src) & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_full(src, nbits);
@@ -258,7 +261,7 @@ static inline int bitmap_full(const unsigned long *src, int nbits)
 
 static inline int bitmap_weight(const unsigned long *src, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return hweight_long(*src & BITMAP_LAST_WORD_MASK(nbits));
 	return __bitmap_weight(src, nbits);
 }
@@ -266,7 +269,7 @@ static inline int bitmap_weight(const unsigned long *src, int nbits)
 static inline void bitmap_shift_right(unsigned long *dst,
 			const unsigned long *src, int n, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src >> n;
 	else
 		__bitmap_shift_right(dst, src, n, nbits);
@@ -275,7 +278,7 @@ static inline void bitmap_shift_right(unsigned long *dst,
 static inline void bitmap_shift_left(unsigned long *dst,
 			const unsigned long *src, int n, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = (*src << n) & BITMAP_LAST_WORD_MASK(nbits);
 	else
 		__bitmap_shift_left(dst, src, n, nbits);
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index d4bf52603e6b..7c178a6baae3 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -268,6 +268,25 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 	bitmap_shift_left(dstp->bits, srcp->bits, n, nbits);
 }
 
+/**
+ * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
+ * @bitmap: the bitmap
+ *
+ * There are a few places where cpumask_var_t isn't appropriate and
+ * static cpumasks must be used (eg. very early boot), yet we don't
+ * expose the definition of 'struct cpumask'.
+ *
+ * This does the conversion, and can be used as a constant initializer.
+ */
+#define to_cpumask(bitmap)						\
+	((struct cpumask *)(1 ? (bitmap)				\
+			    : (void *)sizeof(__check_is_bitmap(bitmap))))
+
+static inline int __check_is_bitmap(const unsigned long *bitmap)
+{
+	return 1;
+}
+
 /*
  * Special-case data structure for "single bit set only" constant CPU masks.
  *
@@ -278,11 +297,11 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 extern const unsigned long
 	cpu_bit_bitmap[BITS_PER_LONG+1][BITS_TO_LONGS(NR_CPUS)];
 
-static inline const cpumask_t *get_cpu_mask(unsigned int cpu)
+static inline const struct cpumask *get_cpu_mask(unsigned int cpu)
 {
 	const unsigned long *p = cpu_bit_bitmap[1 + cpu % BITS_PER_LONG];
 	p -= cpu / BITS_PER_LONG;
-	return (const cpumask_t *)p;
+	return to_cpumask(p);
 }
 
 /*
@@ -416,74 +435,63 @@ int __next_cpu_nr(int n, const cpumask_t *srcp);
 
 /*
  * The following particular system cpumasks and operations manage
- * possible, present, active and online cpus.  Each of them is a fixed size
- * bitmap of size NR_CPUS.
+ * possible, present, active and online cpus.
+ *
+ *     cpu_possible_mask- has bit 'cpu' set iff cpu is populatable
+ *     cpu_present_mask - has bit 'cpu' set iff cpu is populated
+ *     cpu_online_mask  - has bit 'cpu' set iff cpu available to scheduler
+ *     cpu_active_mask  - has bit 'cpu' set iff cpu available to migration
  *
- *  #ifdef CONFIG_HOTPLUG_CPU
- *     cpu_possible_map - has bit 'cpu' set iff cpu is populatable
- *     cpu_present_map  - has bit 'cpu' set iff cpu is populated
- *     cpu_online_map   - has bit 'cpu' set iff cpu available to scheduler
- *     cpu_active_map   - has bit 'cpu' set iff cpu available to migration
- *  #else
- *     cpu_possible_map - has bit 'cpu' set iff cpu is populated
- *     cpu_present_map  - copy of cpu_possible_map
- *     cpu_online_map   - has bit 'cpu' set iff cpu available to scheduler
- *  #endif
+ *  If !CONFIG_HOTPLUG_CPU, present == possible, and active == online.
  *
- *  In either case, NR_CPUS is fixed at compile time, as the static
- *  size of these bitmaps.  The cpu_possible_map is fixed at boot
- *  time, as the set of CPU id's that it is possible might ever
- *  be plugged in at anytime during the life of that system boot.
- *  The cpu_present_map is dynamic(*), representing which CPUs
- *  are currently plugged in.  And cpu_online_map is the dynamic
- *  subset of cpu_present_map, indicating those CPUs available
- *  for scheduling.
+ *  The cpu_possible_mask is fixed at boot time, as the set of CPU id's
+ *  that it is possible might ever be plugged in at anytime during the
+ *  life of that system boot.  The cpu_present_mask is dynamic(*),
+ *  representing which CPUs are currently plugged in.  And
+ *  cpu_online_mask is the dynamic subset of cpu_present_mask,
+ *  indicating those CPUs available for scheduling.
  *
- *  If HOTPLUG is enabled, then cpu_possible_map is forced to have
+ *  If HOTPLUG is enabled, then cpu_possible_mask is forced to have
  *  all NR_CPUS bits set, otherwise it is just the set of CPUs that
  *  ACPI reports present at boot.
  *
- *  If HOTPLUG is enabled, then cpu_present_map varies dynamically,
+ *  If HOTPLUG is enabled, then cpu_present_mask varies dynamically,
  *  depending on what ACPI reports as currently plugged in, otherwise
- *  cpu_present_map is just a copy of cpu_possible_map.
+ *  cpu_present_mask is just a copy of cpu_possible_mask.
  *
- *  (*) Well, cpu_present_map is dynamic in the hotplug case.  If not
- *      hotplug, it's a copy of cpu_possible_map, hence fixed at boot.
+ *  (*) Well, cpu_present_mask is dynamic in the hotplug case.  If not
+ *      hotplug, it's a copy of cpu_possible_mask, hence fixed at boot.
  *
  * Subtleties:
  * 1) UP arch's (NR_CPUS == 1, CONFIG_SMP not defined) hardcode
  *    assumption that their single CPU is online.  The UP
- *    cpu_{online,possible,present}_maps are placebos.  Changing them
+ *    cpu_{online,possible,present}_masks are placebos.  Changing them
  *    will have no useful affect on the following num_*_cpus()
  *    and cpu_*() macros in the UP case.  This ugliness is a UP
  *    optimization - don't waste any instructions or memory references
  *    asking if you're online or how many CPUs there are if there is
  *    only one CPU.
- * 2) Most SMP arch's #define some of these maps to be some
- *    other map specific to that arch.  Therefore, the following
- *    must be #define macros, not inlines.  To see why, examine
- *    the assembly code produced by the following.  Note that
- *    set1() writes phys_x_map, but set2() writes x_map:
- *        int x_map, phys_x_map;
- *        #define set1(a) x_map = a
- *        inline void set2(int a) { x_map = a; }
- *        #define x_map phys_x_map
- *        main(){ set1(3); set2(5); }
  */
 
-extern cpumask_t cpu_possible_map;
-extern cpumask_t cpu_online_map;
-extern cpumask_t cpu_present_map;
-extern cpumask_t cpu_active_map;
+extern const struct cpumask *const cpu_possible_mask;
+extern const struct cpumask *const cpu_online_mask;
+extern const struct cpumask *const cpu_present_mask;
+extern const struct cpumask *const cpu_active_mask;
+
+/* These strip const, as traditionally they weren't const. */
+#define cpu_possible_map	(*(cpumask_t *)cpu_possible_mask)
+#define cpu_online_map		(*(cpumask_t *)cpu_online_mask)
+#define cpu_present_map		(*(cpumask_t *)cpu_present_mask)
+#define cpu_active_map		(*(cpumask_t *)cpu_active_mask)
 
 #if NR_CPUS > 1
-#define num_online_cpus()	cpus_weight_nr(cpu_online_map)
-#define num_possible_cpus()	cpus_weight_nr(cpu_possible_map)
-#define num_present_cpus()	cpus_weight_nr(cpu_present_map)
-#define cpu_online(cpu)		cpu_isset((cpu), cpu_online_map)
-#define cpu_possible(cpu)	cpu_isset((cpu), cpu_possible_map)
-#define cpu_present(cpu)	cpu_isset((cpu), cpu_present_map)
-#define cpu_active(cpu)		cpu_isset((cpu), cpu_active_map)
+#define num_online_cpus()	cpumask_weight(cpu_online_mask)
+#define num_possible_cpus()	cpumask_weight(cpu_possible_mask)
+#define num_present_cpus()	cpumask_weight(cpu_present_mask)
+#define cpu_online(cpu)		cpumask_test_cpu((cpu), cpu_online_mask)
+#define cpu_possible(cpu)	cpumask_test_cpu((cpu), cpu_possible_mask)
+#define cpu_present(cpu)	cpumask_test_cpu((cpu), cpu_present_mask)
+#define cpu_active(cpu)		cpumask_test_cpu((cpu), cpu_active_mask)
 #else
 #define num_online_cpus()	1
 #define num_possible_cpus()	1
@@ -496,10 +504,6 @@ extern cpumask_t cpu_active_map;
 
 #define cpu_is_offline(cpu)	unlikely(!cpu_online(cpu))
 
-#define for_each_possible_cpu(cpu) for_each_cpu_mask_nr((cpu), cpu_possible_map)
-#define for_each_online_cpu(cpu)   for_each_cpu_mask_nr((cpu), cpu_online_map)
-#define for_each_present_cpu(cpu)  for_each_cpu_mask_nr((cpu), cpu_present_map)
-
 /* These are the new versions of the cpumask operators: passed by pointer.
  * The older versions will be implemented in terms of these, then deleted. */
 #define cpumask_bits(maskp) ((maskp)->bits)
@@ -687,7 +691,7 @@ static inline void cpumask_clear_cpu(int cpu, struct cpumask *dstp)
  * No static inline type checking - see Subtlety (1) above.
  */
 #define cpumask_test_cpu(cpu, cpumask) \
-	test_bit(cpumask_check(cpu), (cpumask)->bits)
+	test_bit(cpumask_check(cpu), cpumask_bits((cpumask)))
 
 /**
  * cpumask_test_and_set_cpu - atomically test and set a cpu in a cpumask
@@ -930,7 +934,7 @@ static inline void cpumask_copy(struct cpumask *dstp,
 static inline int cpumask_scnprintf(char *buf, int len,
 				    const struct cpumask *srcp)
 {
-	return bitmap_scnprintf(buf, len, srcp->bits, nr_cpumask_bits);
+	return bitmap_scnprintf(buf, len, cpumask_bits(srcp), nr_cpumask_bits);
 }
 
 /**
@@ -944,7 +948,7 @@ static inline int cpumask_scnprintf(char *buf, int len,
 static inline int cpumask_parse_user(const char __user *buf, int len,
 				     struct cpumask *dstp)
 {
-	return bitmap_parse_user(buf, len, dstp->bits, nr_cpumask_bits);
+	return bitmap_parse_user(buf, len, cpumask_bits(dstp), nr_cpumask_bits);
 }
 
 /**
@@ -959,7 +963,8 @@ static inline int cpumask_parse_user(const char __user *buf, int len,
 static inline int cpulist_scnprintf(char *buf, int len,
 				    const struct cpumask *srcp)
 {
-	return bitmap_scnlistprintf(buf, len, srcp->bits, nr_cpumask_bits);
+	return bitmap_scnlistprintf(buf, len, cpumask_bits(srcp),
+				    nr_cpumask_bits);
 }
 
 /**
@@ -972,26 +977,7 @@ static inline int cpulist_scnprintf(char *buf, int len,
  */
 static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
 {
-	return bitmap_parselist(buf, dstp->bits, nr_cpumask_bits);
-}
-
-/**
- * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
- * @bitmap: the bitmap
- *
- * There are a few places where cpumask_var_t isn't appropriate and
- * static cpumasks must be used (eg. very early boot), yet we don't
- * expose the definition of 'struct cpumask'.
- *
- * This does the conversion, and can be used as a constant initializer.
- */
-#define to_cpumask(bitmap)						\
-	((struct cpumask *)(1 ? (bitmap)				\
-			    : (void *)sizeof(__check_is_bitmap(bitmap))))
-
-static inline int __check_is_bitmap(const unsigned long *bitmap)
-{
-	return 1;
+	return bitmap_parselist(buf, cpumask_bits(dstp), nr_cpumask_bits);
 }
 
 /**
@@ -1025,6 +1011,7 @@ static inline size_t cpumask_size(void)
 #ifdef CONFIG_CPUMASK_OFFSTACK
 typedef struct cpumask *cpumask_var_t;
 
+bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node);
 bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags);
 void alloc_bootmem_cpumask_var(cpumask_var_t *mask);
 void free_cpumask_var(cpumask_var_t mask);
@@ -1038,6 +1025,12 @@ static inline bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 	return true;
 }
 
+static inline bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags,
+					  int node)
+{
+	return true;
+}
+
 static inline void alloc_bootmem_cpumask_var(cpumask_var_t *mask)
 {
 }
@@ -1051,12 +1044,6 @@ static inline void free_bootmem_cpumask_var(cpumask_var_t mask)
 }
 #endif /* CONFIG_CPUMASK_OFFSTACK */
 
-/* The pointer versions of the maps, these will become the primary versions. */
-#define cpu_possible_mask ((const struct cpumask *)&cpu_possible_map)
-#define cpu_online_mask ((const struct cpumask *)&cpu_online_map)
-#define cpu_present_mask ((const struct cpumask *)&cpu_present_map)
-#define cpu_active_mask ((const struct cpumask *)&cpu_active_map)
-
 /* It's common to want to use cpu_all_mask in struct member initializers,
  * so it has to refer to an address rather than a pointer. */
 extern const DECLARE_BITMAP(cpu_all_bits, NR_CPUS);
@@ -1065,51 +1052,16 @@ extern const DECLARE_BITMAP(cpu_all_bits, NR_CPUS);
 /* First bits of cpu_bit_bitmap are in fact unset. */
 #define cpu_none_mask to_cpumask(cpu_bit_bitmap[0])
 
-/* Wrappers for arch boot code to manipulate normally-constant masks */
-static inline void set_cpu_possible(unsigned int cpu, bool possible)
-{
-	if (possible)
-		cpumask_set_cpu(cpu, &cpu_possible_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_possible_map);
-}
+#define for_each_possible_cpu(cpu) for_each_cpu((cpu), cpu_possible_mask)
+#define for_each_online_cpu(cpu)   for_each_cpu((cpu), cpu_online_mask)
+#define for_each_present_cpu(cpu)  for_each_cpu((cpu), cpu_present_mask)
 
-static inline void set_cpu_present(unsigned int cpu, bool present)
-{
-	if (present)
-		cpumask_set_cpu(cpu, &cpu_present_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_present_map);
-}
-
-static inline void set_cpu_online(unsigned int cpu, bool online)
-{
-	if (online)
-		cpumask_set_cpu(cpu, &cpu_online_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_online_map);
-}
-
-static inline void set_cpu_active(unsigned int cpu, bool active)
-{
-	if (active)
-		cpumask_set_cpu(cpu, &cpu_active_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_active_map);
-}
-
-static inline void init_cpu_present(const struct cpumask *src)
-{
-	cpumask_copy(&cpu_present_map, src);
-}
-
-static inline void init_cpu_possible(const struct cpumask *src)
-{
-	cpumask_copy(&cpu_possible_map, src);
-}
-
-static inline void init_cpu_online(const struct cpumask *src)
-{
-	cpumask_copy(&cpu_online_map, src);
-}
+/* Wrappers for arch boot code to manipulate normally-constant masks */
+void set_cpu_possible(unsigned int cpu, bool possible);
+void set_cpu_present(unsigned int cpu, bool present);
+void set_cpu_online(unsigned int cpu, bool online);
+void set_cpu_active(unsigned int cpu, bool active);
+void init_cpu_present(const struct cpumask *src);
+void init_cpu_possible(const struct cpumask *src);
+void init_cpu_online(const struct cpumask *src);
 #endif /* __LINUX_CPUMASK_H */
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index b3dfa72f13b9..40ea5058c2ec 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -50,10 +50,11 @@ int seq_path(struct seq_file *, struct path *, char *);
 int seq_dentry(struct seq_file *, struct dentry *, char *);
 int seq_path_root(struct seq_file *m, struct path *path, struct path *root,
 		  char *esc);
-int seq_bitmap(struct seq_file *m, unsigned long *bits, unsigned int nr_bits);
-static inline int seq_cpumask(struct seq_file *m, cpumask_t *mask)
+int seq_bitmap(struct seq_file *m, const unsigned long *bits,
+				   unsigned int nr_bits);
+static inline int seq_cpumask(struct seq_file *m, const struct cpumask *mask)
 {
-	return seq_bitmap(m, mask->bits, NR_CPUS);
+	return seq_bitmap(m, mask->bits, nr_cpu_ids);
 }
 
 static inline int seq_nodemask(struct seq_file *m, nodemask_t *mask)
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 6e7ba16ff454..b82466968101 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -21,6 +21,9 @@ struct call_single_data {
 	u16 priv;
 };
 
+/* total number of cpus in this system (may exceed NR_CPUS) */
+extern unsigned int total_cpus;
+
 #ifdef CONFIG_SMP
 
 #include <linux/preempt.h>
@@ -64,15 +67,16 @@ extern void smp_cpus_done(unsigned int max_cpus);
  * Call a function on all other processors
  */
 int smp_call_function(void(*func)(void *info), void *info, int wait);
-/* Deprecated: use smp_call_function_many() which uses a cpumask ptr. */
-int smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
-				int wait);
+void smp_call_function_many(const struct cpumask *mask,
+			    void (*func)(void *info), void *info, bool wait);
 
-static inline void smp_call_function_many(const struct cpumask *mask,
-					  void (*func)(void *info), void *info,
-					  int wait)
+/* Deprecated: Use smp_call_function_many which takes a pointer to the mask. */
+static inline int
+smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
+		       int wait)
 {
-	smp_call_function_mask(*mask, func, info, wait);
+	smp_call_function_many(&mask, func, info, wait);
+	return 0;
 }
 
 int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
diff --git a/include/linux/threads.h b/include/linux/threads.h
index 38d1a5d6568e..052b12bec8bd 100644
--- a/include/linux/threads.h
+++ b/include/linux/threads.h
@@ -8,17 +8,17 @@
  */
 
 /*
- * Maximum supported processors that can run under SMP.  This value is
- * set via configure setting.  The maximum is equal to the size of the
- * bitmasks used on that platform, i.e. 32 or 64.  Setting this smaller
- * saves quite a bit of memory.
+ * Maximum supported processors.  Setting this smaller saves quite a
+ * bit of memory.  Use nr_cpu_ids instead of this except for static bitmaps.
  */
-#ifdef CONFIG_SMP
-#define NR_CPUS		CONFIG_NR_CPUS
-#else
-#define NR_CPUS		1
+#ifndef CONFIG_NR_CPUS
+/* FIXME: This should be fixed in the arch's Kconfig */
+#define CONFIG_NR_CPUS	1
 #endif
 
+/* Places which use this should consider cpumask_var_t. */
+#define NR_CPUS		CONFIG_NR_CPUS
+
 #define MIN_THREADS_LEFT_FOR_ROOT 4
 
 /*
diff --git a/kernel/cpu.c b/kernel/cpu.c
index bae131a1211b..2c9f78f3a2fc 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -15,30 +15,8 @@
 #include <linux/stop_machine.h>
 #include <linux/mutex.h>
 
-/*
- * Represents all cpu's present in the system
- * In systems capable of hotplug, this map could dynamically grow
- * as new cpu's are detected in the system via any platform specific
- * method, such as ACPI for e.g.
- */
-cpumask_t cpu_present_map __read_mostly;
-EXPORT_SYMBOL(cpu_present_map);
-
-/*
- * Represents all cpu's that are currently online.
- */
-cpumask_t cpu_online_map __read_mostly;
-EXPORT_SYMBOL(cpu_online_map);
-
-#ifdef CONFIG_INIT_ALL_POSSIBLE
-cpumask_t cpu_possible_map __read_mostly = CPU_MASK_ALL;
-#else
-cpumask_t cpu_possible_map __read_mostly;
-#endif
-EXPORT_SYMBOL(cpu_possible_map);
-
 #ifdef CONFIG_SMP
-/* Serializes the updates to cpu_online_map, cpu_present_map */
+/* Serializes the updates to cpu_online_mask, cpu_present_mask */
 static DEFINE_MUTEX(cpu_add_remove_lock);
 
 static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain);
@@ -65,8 +43,6 @@ void __init cpu_hotplug_init(void)
 	cpu_hotplug.refcount = 0;
 }
 
-cpumask_t cpu_active_map;
-
 #ifdef CONFIG_HOTPLUG_CPU
 
 void get_online_cpus(void)
@@ -97,7 +73,7 @@ EXPORT_SYMBOL_GPL(put_online_cpus);
 
 /*
  * The following two API's must be used when attempting
- * to serialize the updates to cpu_online_map, cpu_present_map.
+ * to serialize the updates to cpu_online_mask, cpu_present_mask.
  */
 void cpu_maps_update_begin(void)
 {
@@ -503,3 +479,71 @@ EXPORT_SYMBOL_GPL(cpu_bit_bitmap);
 
 const DECLARE_BITMAP(cpu_all_bits, NR_CPUS) = CPU_BITS_ALL;
 EXPORT_SYMBOL(cpu_all_bits);
+
+#ifdef CONFIG_INIT_ALL_POSSIBLE
+static DECLARE_BITMAP(cpu_possible_bits, CONFIG_NR_CPUS) __read_mostly
+	= CPU_BITS_ALL;
+#else
+static DECLARE_BITMAP(cpu_possible_bits, CONFIG_NR_CPUS) __read_mostly;
+#endif
+const struct cpumask *const cpu_possible_mask = to_cpumask(cpu_possible_bits);
+EXPORT_SYMBOL(cpu_possible_mask);
+
+static DECLARE_BITMAP(cpu_online_bits, CONFIG_NR_CPUS) __read_mostly;
+const struct cpumask *const cpu_online_mask = to_cpumask(cpu_online_bits);
+EXPORT_SYMBOL(cpu_online_mask);
+
+static DECLARE_BITMAP(cpu_present_bits, CONFIG_NR_CPUS) __read_mostly;
+const struct cpumask *const cpu_present_mask = to_cpumask(cpu_present_bits);
+EXPORT_SYMBOL(cpu_present_mask);
+
+static DECLARE_BITMAP(cpu_active_bits, CONFIG_NR_CPUS) __read_mostly;
+const struct cpumask *const cpu_active_mask = to_cpumask(cpu_active_bits);
+EXPORT_SYMBOL(cpu_active_mask);
+
+void set_cpu_possible(unsigned int cpu, bool possible)
+{
+	if (possible)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_possible_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_possible_bits));
+}
+
+void set_cpu_present(unsigned int cpu, bool present)
+{
+	if (present)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_present_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_present_bits));
+}
+
+void set_cpu_online(unsigned int cpu, bool online)
+{
+	if (online)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_online_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_online_bits));
+}
+
+void set_cpu_active(unsigned int cpu, bool active)
+{
+	if (active)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_active_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_active_bits));
+}
+
+void init_cpu_present(const struct cpumask *src)
+{
+	cpumask_copy(to_cpumask(cpu_present_bits), src);
+}
+
+void init_cpu_possible(const struct cpumask *src)
+{
+	cpumask_copy(to_cpumask(cpu_possible_bits), src);
+}
+
+void init_cpu_online(const struct cpumask *src)
+{
+	cpumask_copy(to_cpumask(cpu_online_bits), src);
+}
diff --git a/kernel/sched.c b/kernel/sched.c
index 1f6b5f05fb65..fea79d7f3ee9 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7222,8 +7222,7 @@ cpu_to_phys_group(int cpu, const struct cpumask *cpu_map,
 {
 	int group;
 #ifdef CONFIG_SCHED_MC
-	/* FIXME: Use cpu_coregroup_mask. */
-	*mask = cpu_coregroup_map(cpu);
+	*mask = *cpu_coregroup_mask(cpu);
 	cpus_and(*mask, *mask, *cpu_map);
 	group = cpumask_first(mask);
 #elif defined(CONFIG_SCHED_SMT)
@@ -7563,7 +7562,7 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		sd = &per_cpu(core_domains, i).sd;
 		SD_INIT(sd, MC);
 		set_domain_attribute(sd, attr);
-		*sched_domain_span(sd) = cpu_coregroup_map(i);
+		*sched_domain_span(sd) = *cpu_coregroup_mask(i);
 		cpumask_and(sched_domain_span(sd),
 			    sched_domain_span(sd), cpu_map);
 		sd->parent = p;
@@ -7601,8 +7600,7 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 #ifdef CONFIG_SCHED_MC
 	/* Set up multi-core groups */
 	for_each_cpu(i, cpu_map) {
-		/* FIXME: Use cpu_coregroup_mask */
-		*this_core_map = cpu_coregroup_map(i);
+		*this_core_map = *cpu_coregroup_mask(i);
 		cpus_and(*this_core_map, *this_core_map, *cpu_map);
 		if (i != cpumask_first(this_core_map))
 			continue;
diff --git a/kernel/smp.c b/kernel/smp.c
index 75c8dde58c55..172b18268909 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -24,8 +24,8 @@ struct call_function_data {
 	struct call_single_data csd;
 	spinlock_t lock;
 	unsigned int refs;
-	cpumask_t cpumask;
 	struct rcu_head rcu_head;
+	unsigned long cpumask_bits[];
 };
 
 struct call_single_queue {
@@ -110,13 +110,13 @@ void generic_smp_call_function_interrupt(void)
 	list_for_each_entry_rcu(data, &call_function_queue, csd.list) {
 		int refs;
 
-		if (!cpu_isset(cpu, data->cpumask))
+		if (!cpumask_test_cpu(cpu, to_cpumask(data->cpumask_bits)))
 			continue;
 
 		data->csd.func(data->csd.info);
 
 		spin_lock(&data->lock);
-		cpu_clear(cpu, data->cpumask);
+		cpumask_clear_cpu(cpu, to_cpumask(data->cpumask_bits));
 		WARN_ON(data->refs == 0);
 		data->refs--;
 		refs = data->refs;
@@ -266,51 +266,19 @@ void __smp_call_function_single(int cpu, struct call_single_data *data)
 	generic_exec_single(cpu, data);
 }
 
-/* Dummy function */
-static void quiesce_dummy(void *unused)
-{
-}
-
-/*
- * Ensure stack based data used in call function mask is safe to free.
- *
- * This is needed by smp_call_function_mask when using on-stack data, because
- * a single call function queue is shared by all CPUs, and any CPU may pick up
- * the data item on the queue at any time before it is deleted. So we need to
- * ensure that all CPUs have transitioned through a quiescent state after
- * this call.
- *
- * This is a very slow function, implemented by sending synchronous IPIs to
- * all possible CPUs. For this reason, we have to alloc data rather than use
- * stack based data even in the case of synchronous calls. The stack based
- * data is then just used for deadlock/oom fallback which will be very rare.
- *
- * If a faster scheme can be made, we could go back to preferring stack based
- * data -- the data allocation/free is non-zero cost.
- */
-static void smp_call_function_mask_quiesce_stack(cpumask_t mask)
-{
-	struct call_single_data data;
-	int cpu;
-
-	data.func = quiesce_dummy;
-	data.info = NULL;
-
-	for_each_cpu_mask(cpu, mask) {
-		data.flags = CSD_FLAG_WAIT;
-		generic_exec_single(cpu, &data);
-	}
-}
+/* FIXME: Shim for archs using old arch_send_call_function_ipi API. */
+#ifndef arch_send_call_function_ipi_mask
+#define arch_send_call_function_ipi_mask(maskp) \
+	arch_send_call_function_ipi(*(maskp))
+#endif
 
 /**
- * smp_call_function_mask(): Run a function on a set of other CPUs.
- * @mask: The set of cpus to run on.
+ * smp_call_function_many(): Run a function on a set of other CPUs.
+ * @mask: The set of cpus to run on (only runs on online subset).
  * @func: The function to run. This must be fast and non-blocking.
  * @info: An arbitrary pointer to pass to the function.
  * @wait: If true, wait (atomically) until function has completed on other CPUs.
  *
- * Returns 0 on success, else a negative status code.
- *
  * If @wait is true, then returns once @func has returned. Note that @wait
  * will be implicitly turned on in case of allocation failures, since
  * we fall back to on-stack allocation.
@@ -319,53 +287,57 @@ static void smp_call_function_mask_quiesce_stack(cpumask_t mask)
  * hardware interrupt handler or from a bottom half handler. Preemption
  * must be disabled when calling this function.
  */
-int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
-			   int wait)
+void smp_call_function_many(const struct cpumask *mask,
+			    void (*func)(void *), void *info,
+			    bool wait)
 {
-	struct call_function_data d;
-	struct call_function_data *data = NULL;
-	cpumask_t allbutself;
+	struct call_function_data *data;
 	unsigned long flags;
-	int cpu, num_cpus;
-	int slowpath = 0;
+	int cpu, next_cpu;
 
 	/* Can deadlock when called with interrupts disabled */
 	WARN_ON(irqs_disabled());
 
-	cpu = smp_processor_id();
-	allbutself = cpu_online_map;
-	cpu_clear(cpu, allbutself);
-	cpus_and(mask, mask, allbutself);
-	num_cpus = cpus_weight(mask);
-
-	/*
-	 * If zero CPUs, return. If just a single CPU, turn this request
-	 * into a targetted single call instead since it's faster.
-	 */
-	if (!num_cpus)
-		return 0;
-	else if (num_cpus == 1) {
-		cpu = first_cpu(mask);
-		return smp_call_function_single(cpu, func, info, wait);
+	/* So, what's a CPU they want?  Ignoring this one. */
+	cpu = cpumask_first_and(mask, cpu_online_mask);
+	if (cpu == smp_processor_id())
+		cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
+	/* No online cpus?  We're done. */
+	if (cpu >= nr_cpu_ids)
+		return;
+
+	/* Do we have another CPU which isn't us? */
+	next_cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
+	if (next_cpu == smp_processor_id())
+		next_cpu = cpumask_next_and(next_cpu, mask, cpu_online_mask);
+
+	/* Fastpath: do that cpu by itself. */
+	if (next_cpu >= nr_cpu_ids) {
+		smp_call_function_single(cpu, func, info, wait);
+		return;
 	}
 
-	data = kmalloc(sizeof(*data), GFP_ATOMIC);
-	if (data) {
-		data->csd.flags = CSD_FLAG_ALLOC;
-		if (wait)
-			data->csd.flags |= CSD_FLAG_WAIT;
-	} else {
-		data = &d;
-		data->csd.flags = CSD_FLAG_WAIT;
-		wait = 1;
-		slowpath = 1;
+	data = kmalloc(sizeof(*data) + cpumask_size(), GFP_ATOMIC);
+	if (unlikely(!data)) {
+		/* Slow path. */
+		for_each_online_cpu(cpu) {
+			if (cpu == smp_processor_id())
+				continue;
+			if (cpumask_test_cpu(cpu, mask))
+				smp_call_function_single(cpu, func, info, wait);
+		}
+		return;
 	}
 
 	spin_lock_init(&data->lock);
+	data->csd.flags = CSD_FLAG_ALLOC;
+	if (wait)
+		data->csd.flags |= CSD_FLAG_WAIT;
 	data->csd.func = func;
 	data->csd.info = info;
-	data->refs = num_cpus;
-	data->cpumask = mask;
+	cpumask_and(to_cpumask(data->cpumask_bits), mask, cpu_online_mask);
+	cpumask_clear_cpu(smp_processor_id(), to_cpumask(data->cpumask_bits));
+	data->refs = cpumask_weight(to_cpumask(data->cpumask_bits));
 
 	spin_lock_irqsave(&call_function_lock, flags);
 	list_add_tail_rcu(&data->csd.list, &call_function_queue);
@@ -377,18 +349,13 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
 	smp_mb();
 
 	/* Send a message to all CPUs in the map */
-	arch_send_call_function_ipi(mask);
+	arch_send_call_function_ipi_mask(to_cpumask(data->cpumask_bits));
 
 	/* optionally wait for the CPUs to complete */
-	if (wait) {
+	if (wait)
 		csd_flag_wait(&data->csd);
-		if (unlikely(slowpath))
-			smp_call_function_mask_quiesce_stack(mask);
-	}
-
-	return 0;
 }
-EXPORT_SYMBOL(smp_call_function_mask);
+EXPORT_SYMBOL(smp_call_function_many);
 
 /**
  * smp_call_function(): Run a function on all other CPUs.
@@ -396,7 +363,7 @@ EXPORT_SYMBOL(smp_call_function_mask);
  * @info: An arbitrary pointer to pass to the function.
  * @wait: If true, wait (atomically) until function has completed on other CPUs.
  *
- * Returns 0 on success, else a negative status code.
+ * Returns 0.
  *
  * If @wait is true, then returns once @func has returned; otherwise
  * it returns just before the target cpu calls @func. In case of allocation
@@ -407,12 +374,10 @@ EXPORT_SYMBOL(smp_call_function_mask);
  */
 int smp_call_function(void (*func)(void *), void *info, int wait)
 {
-	int ret;
-
 	preempt_disable();
-	ret = smp_call_function_mask(cpu_online_map, func, info, wait);
+	smp_call_function_many(cpu_online_mask, func, info, wait);
 	preempt_enable();
-	return ret;
+	return 0;
 }
 EXPORT_SYMBOL(smp_call_function);
 
diff --git a/lib/cpumask.c b/lib/cpumask.c
index 8d03f22c6ced..a24edf137f41 100644
--- a/lib/cpumask.c
+++ b/lib/cpumask.c
@@ -76,15 +76,28 @@ int cpumask_any_but(const struct cpumask *mask, unsigned int cpu)
 
 /* These are not inline because of header tangles. */
 #ifdef CONFIG_CPUMASK_OFFSTACK
-bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
+/**
+ * alloc_cpumask_var_node - allocate a struct cpumask on a given node
+ * @mask: pointer to cpumask_var_t where the cpumask is returned
+ * @flags: GFP_ flags
+ *
+ * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
+ * a nop returning a constant 1 (in <linux/cpumask.h>)
+ * Returns TRUE if memory allocation succeeded, FALSE otherwise.
+ *
+ * In addition, mask will be NULL if this fails.  Note that gcc is
+ * usually smart enough to know that mask can never be NULL if
+ * CONFIG_CPUMASK_OFFSTACK=n, so does code elimination in that case
+ * too.
+ */
+bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node)
 {
 	if (likely(slab_is_available()))
-		*mask = kmalloc(cpumask_size(), flags);
+		*mask = kmalloc_node(cpumask_size(), flags, node);
 	else {
 #ifdef CONFIG_DEBUG_PER_CPU_MAPS
 		printk(KERN_ERR
 			"=> alloc_cpumask_var: kmalloc not available!\n");
-		dump_stack();
 #endif
 		*mask = NULL;
 	}
@@ -96,19 +109,54 @@ bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 #endif
 	return *mask != NULL;
 }
+EXPORT_SYMBOL(alloc_cpumask_var_node);
+
+/**
+ * alloc_cpumask_var - allocate a struct cpumask
+ * @mask: pointer to cpumask_var_t where the cpumask is returned
+ * @flags: GFP_ flags
+ *
+ * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
+ * a nop returning a constant 1 (in <linux/cpumask.h>).
+ *
+ * See alloc_cpumask_var_node.
+ */
+bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
+{
+	return alloc_cpumask_var_node(mask, flags, numa_node_id());
+}
 EXPORT_SYMBOL(alloc_cpumask_var);
 
+/**
+ * alloc_bootmem_cpumask_var - allocate a struct cpumask from the bootmem arena.
+ * @mask: pointer to cpumask_var_t where the cpumask is returned
+ *
+ * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
+ * a nop returning a constant 1 (in <linux/cpumask.h>)
+ * Either returns an allocated (zero-filled) cpumask, or causes the
+ * system to panic.
+ */
 void __init alloc_bootmem_cpumask_var(cpumask_var_t *mask)
 {
 	*mask = alloc_bootmem(cpumask_size());
 }
 
+/**
+ * free_cpumask_var - frees memory allocated for a struct cpumask.
+ * @mask: cpumask to free
+ *
+ * This is safe on a NULL mask.
+ */
 void free_cpumask_var(cpumask_var_t mask)
 {
 	kfree(mask);
 }
 EXPORT_SYMBOL(free_cpumask_var);
 
+/**
+ * free_bootmem_cpumask_var - frees result of alloc_bootmem_cpumask_var
+ * @mask: cpumask to free
+ */
 void __init free_bootmem_cpumask_var(cpumask_var_t mask)
 {
 	free_bootmem((unsigned long)mask, cpumask_size());