167 files changed, 902 insertions, 759 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/smp.h b/arch/alpha/include/asm/smp.h
index 544c69af8168..547e90951cec 100644
--- a/arch/alpha/include/asm/smp.h
+++ b/arch/alpha/include/asm/smp.h
@@ -45,7 +45,6 @@ extern struct cpuinfo_alpha cpu_data[NR_CPUS];
 #define raw_smp_processor_id()  (current_thread_info()->cpu)
 extern int smp_num_cpus;
-#define cpu_possible_map        cpu_present_map
 extern void arch_send_call_function_single_ipi(int cpu);
 extern void arch_send_call_function_ipi(cpumask_t mask);
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/irq.c b/arch/alpha/kernel/irq.c
index c626a821cdcb..d0f1620007f7 100644
--- a/arch/alpha/kernel/irq.c
+++ b/arch/alpha/kernel/irq.c
@@ -55,7 +55,7 @@ int irq_select_affinity(unsigned int irq)
        last_cpu = cpu;
        irq_desc[irq].affinity = cpumask_of_cpu(cpu);
-        irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu));
+        irq_desc[irq].chip->set_affinity(irq, cpumask_of(cpu));
        return 0;
 }
 #endif /* CONFIG_SMP */
diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c
index 351407e07e71..f238370c907d 100644
--- a/arch/alpha/kernel/process.c
+++ b/arch/alpha/kernel/process.c
@@ -94,6 +94,7 @@ common_shutdown_1(void *generic_ptr)
                flags |= 0x00040000UL; /* "remain halted" */
                *pflags = flags;
                cpu_clear(cpuid, cpu_present_map);
+                cpu_clear(cpuid, cpu_possible_map);
                halt();
        }
 #endif
@@ -120,6 +121,7 @@ common_shutdown_1(void *generic_ptr)
 #ifdef CONFIG_SMP
        /* Wait for the secondaries to halt. */
        cpu_clear(boot_cpuid, cpu_present_map);
+        cpu_clear(boot_cpuid, cpu_possible_map);
        while (cpus_weight(cpu_present_map))
                barrier();
 #endif
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/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c
index cf7da10097bb..d953e510f68d 100644
--- a/arch/alpha/kernel/smp.c
+++ b/arch/alpha/kernel/smp.c
@@ -70,11 +70,6 @@ enum ipi_message_type {
 /* Set to a secondary's cpuid when it comes online.  */
 static int smp_secondary_alive __devinitdata = 0;
-/* Which cpus ids came online.  */
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
 int smp_num_probed;             /* Internal processor count */
 int smp_num_cpus = 1;           /* Number that came online.  */
 EXPORT_SYMBOL(smp_num_cpus);
@@ -440,6 +435,7 @@ setup_smp(void)
                                ((char *)cpubase + i*hwrpb->processor_size);
                        if ((cpu->flags & 0x1cc) == 0x1cc) {
                                smp_num_probed++;
+                                cpu_set(i, cpu_possible_map);
                                cpu_set(i, cpu_present_map);
                                cpu->pal_revision = boot_cpu_palrev;
                        }
@@ -473,6 +469,7 @@ smp_prepare_cpus(unsigned int max_cpus)
        /* Nothing to do on a UP box, or when told not to.  */
        if (smp_num_probed == 1 || max_cpus == 0) {
+                cpu_possible_map = cpumask_of_cpu(boot_cpuid);
                cpu_present_map = cpumask_of_cpu(boot_cpuid);
                printk(KERN_INFO "SMP mode deactivated.\n");
                return;
diff --git a/arch/alpha/kernel/sys_dp264.c b/arch/alpha/kernel/sys_dp264.c
index c71b0fd7a61f..ab44c164d9d4 100644
--- a/arch/alpha/kernel/sys_dp264.c
+++ b/arch/alpha/kernel/sys_dp264.c
@@ -177,19 +177,19 @@ cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity)
 }
 static void
-dp264_set_affinity(unsigned int irq, cpumask_t affinity)
+dp264_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
        spin_lock(&dp264_irq_lock);
-        cpu_set_irq_affinity(irq, affinity);
+        cpu_set_irq_affinity(irq, *affinity);
        tsunami_update_irq_hw(cached_irq_mask);
        spin_unlock(&dp264_irq_lock);
 }
 static void
-clipper_set_affinity(unsigned int irq, cpumask_t affinity)
+clipper_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
        spin_lock(&dp264_irq_lock);
-        cpu_set_irq_affinity(irq - 16, affinity);
+        cpu_set_irq_affinity(irq - 16, *affinity);
        tsunami_update_irq_hw(cached_irq_mask);
        spin_unlock(&dp264_irq_lock);
 }
diff --git a/arch/alpha/kernel/sys_titan.c b/arch/alpha/kernel/sys_titan.c
index 52c91ccc1648..27f840a4ad3d 100644
--- a/arch/alpha/kernel/sys_titan.c
+++ b/arch/alpha/kernel/sys_titan.c
@@ -158,10 +158,10 @@ titan_cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity)
 }
 static void
-titan_set_irq_affinity(unsigned int irq, cpumask_t affinity)
+titan_set_irq_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
        spin_lock(&titan_irq_lock);
-        titan_cpu_set_irq_affinity(irq - 16, affinity);
+        titan_cpu_set_irq_affinity(irq - 16, *affinity);
        titan_update_irq_hw(titan_cached_irq_mask);
        spin_unlock(&titan_irq_lock);
 }
diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 7fc9860a97d7..c6884ba1d5ed 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -109,11 +109,11 @@ static void gic_unmask_irq(unsigned int irq)
 }
 #ifdef CONFIG_SMP
-static void gic_set_cpu(unsigned int irq, cpumask_t mask_val)
+static void gic_set_cpu(unsigned int irq, const struct cpumask *mask_val)
 {
        void __iomem *reg = gic_dist_base(irq) + GIC_DIST_TARGET + (gic_irq(irq) & ~3);
        unsigned int shift = (irq % 4) * 8;
-        unsigned int cpu = first_cpu(mask_val);
+        unsigned int cpu = cpumask_first(mask_val);
        u32 val;
        spin_lock(&irq_controller_lock);
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 2f3eb795fa6e..7141cee1fab7 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -174,7 +174,7 @@ static void route_irq(struct irq_desc *desc, unsigned int irq, unsigned int cpu)
        pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->cpu, cpu);
        spin_lock_irq(&desc->lock);
-        desc->chip->set_affinity(irq, cpumask_of_cpu(cpu));
+        desc->chip->set_affinity(irq, cpumask_of(cpu));
        spin_unlock_irq(&desc->lock);
 }
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 019237d21622..55fa7ff96a3e 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -34,16 +34,6 @@
 #include <asm/ptrace.h>
 /*
- * bitmask of present and online CPUs.
- * The present bitmask indicates that the CPU is physically present.
- * The online bitmask indicates that the CPU is up and running.
- */
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
-/*
 * as from 2.5, kernels no longer have an init_tasks structure
 * so we need some other way of telling a new secondary core
 * where to place its SVC stack
diff --git a/arch/arm/mach-at91/at91rm9200_time.c b/arch/arm/mach-at91/at91rm9200_time.c
index d140eae53ded..1ff1bda0a894 100644
--- a/arch/arm/mach-at91/at91rm9200_time.c
+++ b/arch/arm/mach-at91/at91rm9200_time.c
@@ -178,7 +178,6 @@ static struct clock_event_device clkevt = {
        .features       = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
        .shift          = 32,
        .rating         = 150,
-        .cpumask        = CPU_MASK_CPU0,
        .set_next_event = clkevt32k_next_event,
        .set_mode       = clkevt32k_mode,
 };
@@ -206,7 +205,7 @@ void __init at91rm9200_timer_init(void)
        clkevt.mult = div_sc(AT91_SLOW_CLOCK, NSEC_PER_SEC, clkevt.shift);
        clkevt.max_delta_ns = clockevent_delta2ns(AT91_ST_ALMV, &clkevt);
        clkevt.min_delta_ns = clockevent_delta2ns(2, &clkevt) + 1;
-        clkevt.cpumask = cpumask_of_cpu(0);
+        clkevt.cpumask = cpumask_of(0);
        clockevents_register_device(&clkevt);
        /* register clocksource */
diff --git a/arch/arm/mach-at91/at91sam926x_time.c b/arch/arm/mach-at91/at91sam926x_time.c
index 122fd77ed580..b63e1d5f1bad 100644
--- a/arch/arm/mach-at91/at91sam926x_time.c
+++ b/arch/arm/mach-at91/at91sam926x_time.c
@@ -91,7 +91,6 @@ static struct clock_event_device pit_clkevt = {
        .features       = CLOCK_EVT_FEAT_PERIODIC,
        .shift          = 32,
        .rating         = 100,
-        .cpumask        = CPU_MASK_CPU0,
        .set_mode       = pit_clkevt_mode,
 };
@@ -173,6 +172,7 @@ static void __init at91sam926x_pit_init(void)
        /* Set up and register clockevents */
        pit_clkevt.mult = div_sc(pit_rate, NSEC_PER_SEC, pit_clkevt.shift);
+        pit_clkevt.cpumask = cpumask_of(0);
        clockevents_register_device(&pit_clkevt);
 }
diff --git a/arch/arm/mach-davinci/time.c b/arch/arm/mach-davinci/time.c
index 3b9a296b5c4b..f8bcd29d17a6 100644
--- a/arch/arm/mach-davinci/time.c
+++ b/arch/arm/mach-davinci/time.c
@@ -322,7 +322,7 @@ static void __init davinci_timer_init(void)
        clockevent_davinci.min_delta_ns =
                clockevent_delta2ns(1, &clockevent_davinci);
-        clockevent_davinci.cpumask = cpumask_of_cpu(0);
+        clockevent_davinci.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_davinci);
 }
diff --git a/arch/arm/mach-imx/time.c b/arch/arm/mach-imx/time.c
index a11765f5f23b..aff0ebcfa847 100644
--- a/arch/arm/mach-imx/time.c
+++ b/arch/arm/mach-imx/time.c
@@ -184,7 +184,7 @@ static int __init imx_clockevent_init(unsigned long rate)
        clockevent_imx.min_delta_ns =
                clockevent_delta2ns(0xf, &clockevent_imx);
-        clockevent_imx.cpumask = cpumask_of_cpu(0);
+        clockevent_imx.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_imx);
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 7766f469456b..f4656d2ac8a8 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -487,7 +487,7 @@ static int __init ixp4xx_clockevent_init(void)
                clockevent_delta2ns(0xfffffffe, &clockevent_ixp4xx);
        clockevent_ixp4xx.min_delta_ns =
                clockevent_delta2ns(0xf, &clockevent_ixp4xx);
-        clockevent_ixp4xx.cpumask = cpumask_of_cpu(0);
+        clockevent_ixp4xx.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_ixp4xx);
        return 0;
diff --git a/arch/arm/mach-msm/timer.c b/arch/arm/mach-msm/timer.c
index 345a14cb73c3..444d9c0f5ca6 100644
--- a/arch/arm/mach-msm/timer.c
+++ b/arch/arm/mach-msm/timer.c
@@ -182,7 +182,7 @@ static void __init msm_timer_init(void)
                        clockevent_delta2ns(0xf0000000 >> clock->shift, ce);
                /* 4 gets rounded down to 3 */
                ce->min_delta_ns = clockevent_delta2ns(4, ce);
-                ce->cpumask = cpumask_of_cpu(0);
+                ce->cpumask = cpumask_of(0);
                cs->mult = clocksource_hz2mult(clock->freq, cs->shift);
                res = clocksource_register(cs);
diff --git a/arch/arm/mach-ns9xxx/time-ns9360.c b/arch/arm/mach-ns9xxx/time-ns9360.c
index a63424d083d9..41df69721769 100644
--- a/arch/arm/mach-ns9xxx/time-ns9360.c
+++ b/arch/arm/mach-ns9xxx/time-ns9360.c
@@ -173,7 +173,7 @@ static void __init ns9360_timer_init(void)
        ns9360_clockevent_device.min_delta_ns =
                clockevent_delta2ns(1, &ns9360_clockevent_device);
-        ns9360_clockevent_device.cpumask = cpumask_of_cpu(0);
+        ns9360_clockevent_device.cpumask = cpumask_of(0);
        clockevents_register_device(&ns9360_clockevent_device);
        setup_irq(IRQ_NS9360_TIMER0 + TIMER_CLOCKEVENT,
diff --git a/arch/arm/mach-omap1/time.c b/arch/arm/mach-omap1/time.c
index 2cf7e32bd293..495a32c287b4 100644
--- a/arch/arm/mach-omap1/time.c
+++ b/arch/arm/mach-omap1/time.c
@@ -173,7 +173,7 @@ static __init void omap_init_mpu_timer(unsigned long rate)
        clockevent_mpu_timer1.min_delta_ns =
                clockevent_delta2ns(1, &clockevent_mpu_timer1);
-        clockevent_mpu_timer1.cpumask = cpumask_of_cpu(0);
+        clockevent_mpu_timer1.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_mpu_timer1);
 }
diff --git a/arch/arm/mach-omap1/timer32k.c b/arch/arm/mach-omap1/timer32k.c
index 705367ece174..fd3f7396e162 100644
--- a/arch/arm/mach-omap1/timer32k.c
+++ b/arch/arm/mach-omap1/timer32k.c
@@ -187,7 +187,7 @@ static __init void omap_init_32k_timer(void)
        clockevent_32k_timer.min_delta_ns =
                clockevent_delta2ns(1, &clockevent_32k_timer);
-        clockevent_32k_timer.cpumask = cpumask_of_cpu(0);
+        clockevent_32k_timer.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_32k_timer);
 }
diff --git a/arch/arm/mach-omap2/timer-gp.c b/arch/arm/mach-omap2/timer-gp.c
index 589393bedade..ae6036300f60 100644
--- a/arch/arm/mach-omap2/timer-gp.c
+++ b/arch/arm/mach-omap2/timer-gp.c
@@ -120,7 +120,7 @@ static void __init omap2_gp_clockevent_init(void)
        clockevent_gpt.min_delta_ns =
                clockevent_delta2ns(1, &clockevent_gpt);
-        clockevent_gpt.cpumask = cpumask_of_cpu(0);
+        clockevent_gpt.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_gpt);
 }
diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c
index 001624158519..95656a72268d 100644
--- a/arch/arm/mach-pxa/time.c
+++ b/arch/arm/mach-pxa/time.c
@@ -122,7 +122,6 @@ static struct clock_event_device ckevt_pxa_osmr0 = {
        .features       = CLOCK_EVT_FEAT_ONESHOT,
        .shift          = 32,
        .rating         = 200,
-        .cpumask        = CPU_MASK_CPU0,
        .set_next_event = pxa_osmr0_set_next_event,
        .set_mode       = pxa_osmr0_set_mode,
 };
@@ -163,6 +162,7 @@ static void __init pxa_timer_init(void)
                clockevent_delta2ns(0x7fffffff, &ckevt_pxa_osmr0);
        ckevt_pxa_osmr0.min_delta_ns =
                clockevent_delta2ns(MIN_OSCR_DELTA * 2, &ckevt_pxa_osmr0) + 1;
+        ckevt_pxa_osmr0.cpumask = cpumask_of(0);
        cksrc_pxa_oscr0.mult =
                clocksource_hz2mult(clock_tick_rate, cksrc_pxa_oscr0.shift);
diff --git a/arch/arm/mach-realview/core.c b/arch/arm/mach-realview/core.c
index 5f1d55963ced..bd2aa4f16141 100644
--- a/arch/arm/mach-realview/core.c
+++ b/arch/arm/mach-realview/core.c
@@ -624,7 +624,7 @@ static struct clock_event_device timer0_clockevent =	 {
        .set_mode       = timer_set_mode,
        .set_next_event = timer_set_next_event,
        .rating         = 300,
-        .cpumask        = CPU_MASK_ALL,
+        .cpumask        = cpu_all_mask,
 };
 static void __init realview_clockevents_init(unsigned int timer_irq)
diff --git a/arch/arm/mach-realview/localtimer.c b/arch/arm/mach-realview/localtimer.c
index 9019ef2e5611..67d6d9cc68b2 100644
--- a/arch/arm/mach-realview/localtimer.c
+++ b/arch/arm/mach-realview/localtimer.c
@@ -154,7 +154,7 @@ void __cpuinit local_timer_setup(void)
        clk->set_mode           = local_timer_set_mode;
        clk->set_next_event     = local_timer_set_next_event;
        clk->irq                = IRQ_LOCALTIMER;
-        clk->cpumask            = cpumask_of_cpu(cpu);
+        clk->cpumask            = cpumask_of(cpu);
        clk->shift              = 20;
        clk->mult               = div_sc(mpcore_timer_rate, NSEC_PER_SEC, clk->shift);
        clk->max_delta_ns       = clockevent_delta2ns(0xffffffff, clk);
@@ -193,7 +193,7 @@ void __cpuinit local_timer_setup(void)
        clk->rating             = 200;
        clk->set_mode           = dummy_timer_set_mode;
        clk->broadcast          = smp_timer_broadcast;
-        clk->cpumask            = cpumask_of_cpu(cpu);
+        clk->cpumask            = cpumask_of(cpu);
        clockevents_register_device(clk);
 }
diff --git a/arch/arm/mach-sa1100/time.c b/arch/arm/mach-sa1100/time.c
index 8c5e727f3b75..711c0295c66f 100644
--- a/arch/arm/mach-sa1100/time.c
+++ b/arch/arm/mach-sa1100/time.c
@@ -73,7 +73,6 @@ static struct clock_event_device ckevt_sa1100_osmr0 = {
        .features       = CLOCK_EVT_FEAT_ONESHOT,
        .shift          = 32,
        .rating         = 200,
-        .cpumask        = CPU_MASK_CPU0,
        .set_next_event = sa1100_osmr0_set_next_event,
        .set_mode       = sa1100_osmr0_set_mode,
 };
@@ -110,6 +109,7 @@ static void __init sa1100_timer_init(void)
                clockevent_delta2ns(0x7fffffff, &ckevt_sa1100_osmr0);
        ckevt_sa1100_osmr0.min_delta_ns =
                clockevent_delta2ns(MIN_OSCR_DELTA * 2, &ckevt_sa1100_osmr0) + 1;
+        ckevt_sa1100_osmr0.cpumask = cpumask_of(0);
        cksrc_sa1100_oscr.mult =
                clocksource_hz2mult(CLOCK_TICK_RATE, cksrc_sa1100_oscr.shift);
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index df25aa138509..1c43494f5c42 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -1005,7 +1005,7 @@ static void __init versatile_timer_init(void)
        timer0_clockevent.min_delta_ns =
                clockevent_delta2ns(0xf, &timer0_clockevent);
-        timer0_clockevent.cpumask = cpumask_of_cpu(0);
+        timer0_clockevent.cpumask = cpumask_of(0);
        clockevents_register_device(&timer0_clockevent);
 }
diff --git a/arch/arm/oprofile/op_model_mpcore.c b/arch/arm/oprofile/op_model_mpcore.c
index 4de366e8b4c5..6d6bd5899240 100644
--- a/arch/arm/oprofile/op_model_mpcore.c
+++ b/arch/arm/oprofile/op_model_mpcore.c
@@ -260,10 +260,10 @@ static void em_stop(void)
 static void em_route_irq(int irq, unsigned int cpu)
 {
        struct irq_desc *desc = irq_desc + irq;
-        cpumask_t mask = cpumask_of_cpu(cpu);
+        const struct cpumask *mask = cpumask_of(cpu);
        spin_lock_irq(&desc->lock);
-        desc->affinity = mask;
+        desc->affinity = *mask;
        desc->chip->set_affinity(irq, mask);
        spin_unlock_irq(&desc->lock);
 }
diff --git a/arch/arm/plat-mxc/time.c b/arch/arm/plat-mxc/time.c
index fd28f5194f71..758a1293bcfa 100644
--- a/arch/arm/plat-mxc/time.c
+++ b/arch/arm/plat-mxc/time.c
@@ -190,7 +190,7 @@ static int __init mxc_clockevent_init(void)
        clockevent_mxc.min_delta_ns =
                        clockevent_delta2ns(0xff, &clockevent_mxc);
-        clockevent_mxc.cpumask = cpumask_of_cpu(0);
+        clockevent_mxc.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_mxc);
diff --git a/arch/arm/plat-orion/time.c b/arch/arm/plat-orion/time.c
index 544d6b327f3a..6fa2923e6dca 100644
--- a/arch/arm/plat-orion/time.c
+++ b/arch/arm/plat-orion/time.c
@@ -149,7 +149,6 @@ static struct clock_event_device orion_clkevt = {
        .features       = CLOCK_EVT_FEAT_ONESHOT | CLOCK_EVT_FEAT_PERIODIC,
        .shift          = 32,
        .rating         = 300,
-        .cpumask        = CPU_MASK_CPU0,
        .set_next_event = orion_clkevt_next_event,
        .set_mode       = orion_clkevt_mode,
 };
@@ -199,5 +198,6 @@ void __init orion_time_init(unsigned int irq, unsigned int tclk)
        orion_clkevt.mult = div_sc(tclk, NSEC_PER_SEC, orion_clkevt.shift);
        orion_clkevt.max_delta_ns = clockevent_delta2ns(0xfffffffe, &orion_clkevt);
        orion_clkevt.min_delta_ns = clockevent_delta2ns(1, &orion_clkevt);
+        orion_clkevt.cpumask = cpumask_of(0);
        clockevents_register_device(&orion_clkevt);
 }
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
index 283481d74a5b..0ff46bf873b0 100644
--- a/arch/avr32/kernel/time.c
+++ b/arch/avr32/kernel/time.c
@@ -106,7 +106,6 @@ static struct clock_event_device comparator = {
        .features       = CLOCK_EVT_FEAT_ONESHOT,
        .shift          = 16,
        .rating         = 50,
-        .cpumask        = CPU_MASK_CPU0,
        .set_next_event = comparator_next_event,
        .set_mode       = comparator_mode,
 };
@@ -134,6 +133,7 @@ void __init time_init(void)
        comparator.mult = div_sc(counter_hz, NSEC_PER_SEC, comparator.shift);
        comparator.max_delta_ns = clockevent_delta2ns((u32)~0, &comparator);
        comparator.min_delta_ns = clockevent_delta2ns(50, &comparator) + 1;
+        comparator.cpumask = cpumask_of(0);
        sysreg_write(COMPARE, 0);
        timer_irqaction.dev_id = &comparator;
diff --git a/arch/blackfin/kernel/time-ts.c b/arch/blackfin/kernel/time-ts.c
index e887efc86c29..0ed2badfd746 100644
--- a/arch/blackfin/kernel/time-ts.c
+++ b/arch/blackfin/kernel/time-ts.c
@@ -162,7 +162,6 @@ static struct clock_event_device clockevent_bfin = {
        .name           = "bfin_core_timer",
        .features       = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
        .shift          = 32,
-        .cpumask        = CPU_MASK_CPU0,
        .set_next_event = bfin_timer_set_next_event,
        .set_mode       = bfin_timer_set_mode,
 };
@@ -193,6 +192,7 @@ static int __init bfin_clockevent_init(void)
        clockevent_bfin.mult = div_sc(timer_clk, NSEC_PER_SEC, clockevent_bfin.shift);
        clockevent_bfin.max_delta_ns = clockevent_delta2ns(-1, &clockevent_bfin);
        clockevent_bfin.min_delta_ns = clockevent_delta2ns(100, &clockevent_bfin);
+        clockevent_bfin.cpumask = cpumask_of(0);
        clockevents_register_device(&clockevent_bfin);
        return 0;
diff --git a/arch/cris/arch-v32/kernel/irq.c b/arch/cris/arch-v32/kernel/irq.c
index 173c141ac9ba..295131fee710 100644
--- a/arch/cris/arch-v32/kernel/irq.c
+++ b/arch/cris/arch-v32/kernel/irq.c
@@ -325,11 +325,11 @@ static void end_crisv32_irq(unsigned int irq)
 {
 }
-void set_affinity_crisv32_irq(unsigned int irq, cpumask_t dest)
+void set_affinity_crisv32_irq(unsigned int irq, const struct cpumask *dest)
 {
        unsigned long flags;
        spin_lock_irqsave(&irq_lock, flags);
-        irq_allocations[irq - FIRST_IRQ].mask = dest;
+        irq_allocations[irq - FIRST_IRQ].mask = *dest;
        spin_unlock_irqrestore(&irq_lock, flags);
 }
diff --git a/arch/cris/arch-v32/kernel/smp.c b/arch/cris/arch-v32/kernel/smp.c
index 52e16c6436f9..9dac17334640 100644
--- a/arch/cris/arch-v32/kernel/smp.c
+++ b/arch/cris/arch-v32/kernel/smp.c
@@ -29,11 +29,7 @@
 spinlock_t cris_atomic_locks[] = { [0 ... LOCK_COUNT - 1] = SPIN_LOCK_UNLOCKED};
 /* CPU masks */
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
 cpumask_t phys_cpu_present_map = CPU_MASK_NONE;
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
 EXPORT_SYMBOL(phys_cpu_present_map);
 /* Variables used during SMP boot */
diff --git a/arch/cris/include/asm/smp.h b/arch/cris/include/asm/smp.h
index dba33aba3e95..c615a06dd757 100644
--- a/arch/cris/include/asm/smp.h
+++ b/arch/cris/include/asm/smp.h
@@ -4,7 +4,6 @@
 #include <linux/cpumask.h>
 extern cpumask_t phys_cpu_present_map;
-extern cpumask_t cpu_possible_map;
 #define raw_smp_processor_id() (current_thread_info()->cpu)
diff --git a/arch/ia64/hp/sim/hpsim_irq.c b/arch/ia64/hp/sim/hpsim_irq.c
index c2f58ff364e7..cc0a3182db3c 100644
--- a/arch/ia64/hp/sim/hpsim_irq.c
+++ b/arch/ia64/hp/sim/hpsim_irq.c
@@ -22,7 +22,7 @@ hpsim_irq_noop (unsigned int irq)
 }
 static void
-hpsim_set_affinity_noop (unsigned int a, cpumask_t b)
+hpsim_set_affinity_noop(unsigned int a, const struct cpumask *b)
 {
 }
diff --git a/arch/ia64/include/asm/smp.h b/arch/ia64/include/asm/smp.h
index 12d96e0cd513..21c402365d0e 100644
--- a/arch/ia64/include/asm/smp.h
+++ b/arch/ia64/include/asm/smp.h
@@ -57,7 +57,6 @@ extern struct smp_boot_data {
 extern char no_int_routing __devinitdata;
-extern cpumask_t cpu_online_map;
 extern cpumask_t cpu_core_map[NR_CPUS];
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 extern int smp_num_siblings;
diff --git a/arch/ia64/include/asm/topology.h b/arch/ia64/include/asm/topology.h
index 35bcb641c9e5..66f0f1ef9e7f 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
@@ -121,6 +122,10 @@ 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 5c4674ae8aea..5cfd3d91001a 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -330,25 +330,25 @@ unmask_irq (unsigned int irq)
 static void
-iosapic_set_affinity (unsigned int irq, cpumask_t mask)
+iosapic_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 #ifdef CONFIG_SMP
        u32 high32, low32;
-        int dest, rte_index;
+        int cpu, dest, rte_index;
        int redir = (irq & IA64_IRQ_REDIRECTED) ? 1 : 0;
        struct iosapic_rte_info *rte;
        struct iosapic *iosapic;
        irq &= (~IA64_IRQ_REDIRECTED);
-        cpus_and(mask, mask, cpu_online_map);
+        cpu = cpumask_first_and(cpu_online_mask, mask);
-        if (cpus_empty(mask))
+        if (cpu >= nr_cpu_ids)
                return;
-        if (irq_prepare_move(irq, first_cpu(mask)))
+        if (irq_prepare_move(irq, cpu))
                return;
-        dest = cpu_physical_id(first_cpu(mask));
+        dest = cpu_physical_id(cpu);
        if (!iosapic_intr_info[irq].count)
                return;                 /* not an IOSAPIC interrupt */
@@ -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);
+                cpu_mask = cpumask_of_node(iosapic_lists[iosapic_index].node);
-                cpus_and(cpu_mask, cpu_mask, domain);
+                num_cpus = 0;
-                for_each_cpu_mask(numa_cpu, cpu_mask) {
+                for_each_cpu_and(numa_cpu, cpu_mask, &domain) {
-                        if (!cpu_online(numa_cpu))
+                        if (cpu_online(numa_cpu))
-                                cpu_clear(numa_cpu, cpu_mask);
+                                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++)
+                for_each_cpu_and(numa_cpu, cpu_mask, &domain)
-                        numa_cpu = next_cpu(numa_cpu, cpu_mask);
+                        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/kernel/irq.c b/arch/ia64/kernel/irq.c
index 7fd18f54c056..0b6db53fedcf 100644
--- a/arch/ia64/kernel/irq.c
+++ b/arch/ia64/kernel/irq.c
@@ -133,7 +133,6 @@ unsigned int vectors_in_migration[NR_IRQS];
 */
 static void migrate_irqs(void)
 {
-        cpumask_t       mask;
        irq_desc_t *desc;
        int             irq, new_cpu;
@@ -152,15 +151,14 @@ static void migrate_irqs(void)
                if (desc->status == IRQ_PER_CPU)
                        continue;
-                cpus_and(mask, irq_desc[irq].affinity, cpu_online_map);
+                if (cpumask_any_and(&irq_desc[irq].affinity, cpu_online_mask)
-                if (any_online_cpu(mask) == NR_CPUS) {
+                    >= nr_cpu_ids) {
                        /*
                         * Save it for phase 2 processing
                         */
                        vectors_in_migration[irq] = irq;
                        new_cpu = any_online_cpu(cpu_online_map);
-                        mask = cpumask_of_cpu(new_cpu);
                        /*
                         * Al three are essential, currently WARN_ON.. maybe panic?
@@ -168,7 +166,8 @@ static void migrate_irqs(void)
                        if (desc->chip && desc->chip->disable &&
                                desc->chip->enable && desc->chip->set_affinity) {
                                desc->chip->disable(irq);
-                                desc->chip->set_affinity(irq, mask);
+                                desc->chip->set_affinity(irq,
+                                                         cpumask_of(new_cpu));
                                desc->chip->enable(irq);
                        } else {
                                WARN_ON((!(desc->chip) || !(desc->chip->disable) ||
diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index 702a09c13238..890339339035 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -49,11 +49,12 @@
 static struct irq_chip  ia64_msi_chip;
 #ifdef CONFIG_SMP
-static void ia64_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
+static void ia64_set_msi_irq_affinity(unsigned int irq,
+                                      const cpumask_t *cpu_mask)
 {
        struct msi_msg msg;
        u32 addr, data;
-        int cpu = first_cpu(cpu_mask);
+        int cpu = first_cpu(*cpu_mask);
        if (!cpu_online(cpu))
                return;
@@ -166,12 +167,11 @@ 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_cfg *cfg = irq_cfg + irq;
        struct msi_msg msg;
-        int cpu = first_cpu(mask);
+        int cpu = cpumask_first(mask);
        if (!cpu_online(cpu))
                return;
@@ -187,7 +187,7 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
        msg.address_lo |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu));
        dmar_msi_write(irq, &msg);
-        irq_desc[irq].affinity = mask;
+        irq_desc[irq].affinity = *mask;
 }
 #endif /* CONFIG_SMP */
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 1dcbb85fc4ee..11463994a7d5 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -131,12 +131,6 @@ struct task_struct *task_for_booting_cpu;
 */
 DEFINE_PER_CPU(int, cpu_state);
-/* Bitmasks of currently online, and possible CPUs */
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_possible_map);
 cpumask_t cpu_core_map[NR_CPUS] __cacheline_aligned;
 EXPORT_SYMBOL(cpu_core_map);
 DEFINE_PER_CPU_SHARED_ALIGNED(cpumask_t, cpu_sibling_map);
@@ -688,7 +682,7 @@ int migrate_platform_irqs(unsigned int cpu)
 {
        int new_cpei_cpu;
        irq_desc_t *desc = NULL;
-        cpumask_t       mask;
+        const struct cpumask *mask;
        int             retval = 0;
        /*
@@ -701,7 +695,7 @@ int migrate_platform_irqs(unsigned int cpu)
                         * Now re-target the CPEI to a different processor
                         */
                        new_cpei_cpu = any_online_cpu(cpu_online_map);
-                        mask = cpumask_of_cpu(new_cpei_cpu);
+                        mask = cpumask_of(new_cpei_cpu);
                        set_cpei_target_cpu(new_cpei_cpu);
                        desc = irq_desc + ia64_cpe_irq;
                        /*
diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c
index c75b914f2d6b..a8d61a3e9a94 100644
--- a/arch/ia64/kernel/topology.c
+++ b/arch/ia64/kernel/topology.c
@@ -219,7 +219,7 @@ static ssize_t show_shared_cpu_map(struct cache_info *this_leaf, char *buf)
        cpumask_t shared_cpu_map;
        cpus_and(shared_cpu_map, this_leaf->shared_cpu_map, cpu_online_map);
-        len = cpumask_scnprintf(buf, NR_CPUS+1, shared_cpu_map);
+        len = cpumask_scnprintf(buf, NR_CPUS+1, &shared_cpu_map);
        len += sprintf(buf+len, "\n");
        return len;
 }
diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c
index 0c66dbdd1d72..66fd705e82c0 100644
--- a/arch/ia64/sn/kernel/irq.c
+++ b/arch/ia64/sn/kernel/irq.c
@@ -227,14 +227,14 @@ finish_up:
        return new_irq_info;
 }
-static void sn_set_affinity_irq(unsigned int irq, cpumask_t mask)
+static void sn_set_affinity_irq(unsigned int irq, const struct cpumask *mask)
 {
        struct sn_irq_info *sn_irq_info, *sn_irq_info_safe;
        nasid_t nasid;
        int slice;
-        nasid = cpuid_to_nasid(first_cpu(mask));
+        nasid = cpuid_to_nasid(cpumask_first(mask));
-        slice = cpuid_to_slice(first_cpu(mask));
+        slice = cpuid_to_slice(cpumask_first(mask));
        list_for_each_entry_safe(sn_irq_info, sn_irq_info_safe,
                                 sn_irq_lh[irq], list)
diff --git a/arch/ia64/sn/kernel/msi_sn.c b/arch/ia64/sn/kernel/msi_sn.c
index 83f190ffe350..ca553b0429ce 100644
--- a/arch/ia64/sn/kernel/msi_sn.c
+++ b/arch/ia64/sn/kernel/msi_sn.c
@@ -151,7 +151,8 @@ int sn_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *entry)
 }
 #ifdef CONFIG_SMP
-static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
+static void sn_set_msi_irq_affinity(unsigned int irq,
+                                    const struct cpumask *cpu_mask)
 {
        struct msi_msg msg;
        int slice;
@@ -164,7 +165,7 @@ static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
        struct sn_pcibus_provider *provider;
        unsigned int cpu;
-        cpu = first_cpu(cpu_mask);
+        cpu = cpumask_first(cpu_mask);
        sn_irq_info = sn_msi_info[irq].sn_irq_info;
        if (sn_irq_info == NULL || sn_irq_info->irq_int_bit >= 0)
                return;
@@ -204,7 +205,7 @@ static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
        msg.address_lo = (u32)(bus_addr & 0x00000000ffffffff);
        write_msi_msg(irq, &msg);
-        irq_desc[irq].affinity = cpu_mask;
+        irq_desc[irq].affinity = *cpu_mask;
 }
 #endif /* CONFIG_SMP */
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_cpu_and(i, cpu_online_mask,
-                        for_each_online_cpu(i) {
+                                         cpumask_of_node(ordinal)) {
-                                if (cpu_isset(i, cpumask)) {
+                                slice = 'a' + cpuid_to_slice(i);
-                                        slice = 'a' + cpuid_to_slice(i);
+                                c = cpu_data(i);
-                                        c = cpu_data(i);
+                                seq_printf(s, "cpu %d %s%c local"
-                                        seq_printf(s, "cpu %d %s%c local"
+                                           " freq %luMHz, arch ia64",
-                                                " freq %luMHz, arch ia64",
+                                           i, obj->location, slice,
-                                                i, obj->location, slice,
+                                           c->proc_freq / 1000000);
-                                                c->proc_freq / 1000000);
+                                for_each_online_cpu(j) {
-                                        for_each_online_cpu(j) {
+                                        seq_printf(s, j ? ":%d" : ", dist %d",
-                                                seq_printf(s, j ? ":%d" : ", dist %d",
+                                                   node_distance(
-                                                        node_distance(
                                                        cpu_to_node(i),
                                                        cpu_to_node(j)));
-                                        }
-                                        seq_putc(s, '\n');
                                }
+                                seq_putc(s, '\n');
                        }
                }
        }
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index 29047d5c259a..cabba332cc48 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -10,6 +10,7 @@ config M32R
        default y
        select HAVE_IDE
        select HAVE_OPROFILE
+        select INIT_ALL_POSSIBLE
 config SBUS
        bool
diff --git a/arch/m32r/kernel/smpboot.c b/arch/m32r/kernel/smpboot.c
index 39cb6da72dcb..0f06b3722e96 100644
--- a/arch/m32r/kernel/smpboot.c
+++ b/arch/m32r/kernel/smpboot.c
@@ -73,17 +73,11 @@ static unsigned int bsp_phys_id = -1;
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map;
-/* Bitmask of currently online CPUs */
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
 cpumask_t cpu_bootout_map;
 cpumask_t cpu_bootin_map;
 static cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
 EXPORT_SYMBOL(cpu_callout_map);
-cpumask_t cpu_possible_map = CPU_MASK_ALL;
-EXPORT_SYMBOL(cpu_possible_map);
 /* Per CPU bogomips and other parameters */
 struct cpuinfo_m32r cpu_data[NR_CPUS] __cacheline_aligned;
diff --git a/arch/m68knommu/platform/coldfire/pit.c b/arch/m68knommu/platform/coldfire/pit.c
index c5b916700b22..2a12e7fa9748 100644
--- a/arch/m68knommu/platform/coldfire/pit.c
+++ b/arch/m68knommu/platform/coldfire/pit.c
@@ -156,7 +156,7 @@ void hw_timer_init(void)
 {
        u32 imr;
-        cf_pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+        cf_pit_clockevent.cpumask = cpumask_of(smp_processor_id());
        cf_pit_clockevent.mult = div_sc(FREQ, NSEC_PER_SEC, 32);
        cf_pit_clockevent.max_delta_ns =
                clockevent_delta2ns(0xFFFF, &cf_pit_clockevent);
diff --git a/arch/mips/include/asm/irq.h b/arch/mips/include/asm/irq.h
index a58f0eecc68f..abc62aa744ac 100644
--- a/arch/mips/include/asm/irq.h
+++ b/arch/mips/include/asm/irq.h
@@ -49,7 +49,8 @@ static inline void smtc_im_ack_irq(unsigned int irq)
 #ifdef CONFIG_MIPS_MT_SMTC_IRQAFF
 #include <linux/cpumask.h>
-extern void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity);
+extern void plat_set_irq_affinity(unsigned int irq,
+                                  const struct cpumask *affinity);
 extern void smtc_forward_irq(unsigned int irq);
 /*
diff --git a/arch/mips/include/asm/mach-ip27/topology.h b/arch/mips/include/asm/mach-ip27/topology.h
index 7785bec732f2..c1c3f5b2f18f 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/mips/include/asm/smp.h b/arch/mips/include/asm/smp.h
index 0ff5b523ea77..86557b5d1b3f 100644
--- a/arch/mips/include/asm/smp.h
+++ b/arch/mips/include/asm/smp.h
@@ -38,9 +38,6 @@ extern int __cpu_logical_map[NR_CPUS];
 #define SMP_RESCHEDULE_YOURSELF 0x1     /* XXX braindead */
 #define SMP_CALL_FUNCTION       0x2
-extern cpumask_t phys_cpu_present_map;
-#define cpu_possible_map        phys_cpu_present_map
 extern void asmlinkage smp_bootstrap(void);
 /*
diff --git a/arch/mips/jazz/irq.c b/arch/mips/jazz/irq.c
index d7f8a782aae4..03965cb1b252 100644
--- a/arch/mips/jazz/irq.c
+++ b/arch/mips/jazz/irq.c
@@ -146,7 +146,7 @@ void __init plat_time_init(void)
        BUG_ON(HZ != 100);
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        clockevents_register_device(cd);
        action->dev_id = cd;
        setup_irq(JAZZ_TIMER_IRQ, action);
diff --git a/arch/mips/kernel/cevt-bcm1480.c b/arch/mips/kernel/cevt-bcm1480.c
index 0a57f86945f1..b820661678b0 100644
--- a/arch/mips/kernel/cevt-bcm1480.c
+++ b/arch/mips/kernel/cevt-bcm1480.c
@@ -126,7 +126,7 @@ void __cpuinit sb1480_clockevent_init(void)
        cd->min_delta_ns        = clockevent_delta2ns(2, cd);
        cd->rating              = 200;
        cd->irq                 = irq;
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        cd->set_next_event      = sibyte_next_event;
        cd->set_mode            = sibyte_set_mode;
        clockevents_register_device(cd);
@@ -148,6 +148,6 @@ void __cpuinit sb1480_clockevent_init(void)
        action->name    = name;
        action->dev_id  = cd;
-        irq_set_affinity(irq, cpumask_of_cpu(cpu));
+        irq_set_affinity(irq, cpumask_of(cpu));
        setup_irq(irq, action);
 }
diff --git a/arch/mips/kernel/cevt-ds1287.c b/arch/mips/kernel/cevt-ds1287.c
index df4acb68bfb5..1ada45ea0700 100644
--- a/arch/mips/kernel/cevt-ds1287.c
+++ b/arch/mips/kernel/cevt-ds1287.c
@@ -88,7 +88,6 @@ static void ds1287_event_handler(struct clock_event_device *dev)
 static struct clock_event_device ds1287_clockevent = {
        .name           = "ds1287",
        .features       = CLOCK_EVT_FEAT_PERIODIC,
-        .cpumask        = CPU_MASK_CPU0,
        .set_next_event = ds1287_set_next_event,
        .set_mode       = ds1287_set_mode,
        .event_handler  = ds1287_event_handler,
@@ -122,6 +121,7 @@ int __init ds1287_clockevent_init(int irq)
        clockevent_set_clock(cd, 32768);
        cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
        cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+        cd->cpumask = cpumask_of(0);
        clockevents_register_device(&ds1287_clockevent);
diff --git a/arch/mips/kernel/cevt-gt641xx.c b/arch/mips/kernel/cevt-gt641xx.c
index 6e2f58520afb..e9b787feedcb 100644
--- a/arch/mips/kernel/cevt-gt641xx.c
+++ b/arch/mips/kernel/cevt-gt641xx.c
@@ -96,7 +96,6 @@ static void gt641xx_timer0_event_handler(struct clock_event_device *dev)
 static struct clock_event_device gt641xx_timer0_clockevent = {
        .name           = "gt641xx-timer0",
        .features       = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
-        .cpumask        = CPU_MASK_CPU0,
        .irq            = GT641XX_TIMER0_IRQ,
        .set_next_event = gt641xx_timer0_set_next_event,
        .set_mode       = gt641xx_timer0_set_mode,
@@ -132,6 +131,7 @@ static int __init gt641xx_timer0_clockevent_init(void)
        clockevent_set_clock(cd, gt641xx_base_clock);
        cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
        cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+        cd->cpumask = cpumask_of(0);
        clockevents_register_device(&gt641xx_timer0_clockevent);
diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
index 4a4c59f2737a..e1ec83b68031 100644
--- a/arch/mips/kernel/cevt-r4k.c
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -195,7 +195,7 @@ int __cpuinit mips_clockevent_init(void)
        cd->rating              = 300;
        cd->irq                 = irq;
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        cd->set_next_event      = mips_next_event;
        cd->set_mode            = mips_set_clock_mode;
        cd->event_handler       = mips_event_handler;
diff --git a/arch/mips/kernel/cevt-sb1250.c b/arch/mips/kernel/cevt-sb1250.c
index 63ac3ad462bc..a2eebaafda52 100644
--- a/arch/mips/kernel/cevt-sb1250.c
+++ b/arch/mips/kernel/cevt-sb1250.c
@@ -125,7 +125,7 @@ void __cpuinit sb1250_clockevent_init(void)
        cd->min_delta_ns        = clockevent_delta2ns(2, cd);
        cd->rating              = 200;
        cd->irq                 = irq;
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        cd->set_next_event      = sibyte_next_event;
        cd->set_mode            = sibyte_set_mode;
        clockevents_register_device(cd);
@@ -147,6 +147,6 @@ void __cpuinit sb1250_clockevent_init(void)
        action->name    = name;
        action->dev_id  = cd;
-        irq_set_affinity(irq, cpumask_of_cpu(cpu));
+        irq_set_affinity(irq, cpumask_of(cpu));
        setup_irq(irq, action);
 }
diff --git a/arch/mips/kernel/cevt-smtc.c b/arch/mips/kernel/cevt-smtc.c
index 5162fe4b5952..6d45e24db5bf 100644
--- a/arch/mips/kernel/cevt-smtc.c
+++ b/arch/mips/kernel/cevt-smtc.c
@@ -292,7 +292,7 @@ int __cpuinit mips_clockevent_init(void)
        cd->rating              = 300;
        cd->irq                 = irq;
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        cd->set_next_event      = mips_next_event;
        cd->set_mode            = mips_set_clock_mode;
        cd->event_handler       = mips_event_handler;
diff --git a/arch/mips/kernel/cevt-txx9.c b/arch/mips/kernel/cevt-txx9.c
index b5fc4eb412d2..eccf7d6096bd 100644
--- a/arch/mips/kernel/cevt-txx9.c
+++ b/arch/mips/kernel/cevt-txx9.c
@@ -112,7 +112,6 @@ static struct clock_event_device txx9tmr_clock_event_device = {
        .name           = "TXx9",
        .features       = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
        .rating         = 200,
-        .cpumask        = CPU_MASK_CPU0,
        .set_mode       = txx9tmr_set_mode,
        .set_next_event = txx9tmr_set_next_event,
 };
@@ -150,6 +149,7 @@ void __init txx9_clockevent_init(unsigned long baseaddr, int irq,
                clockevent_delta2ns(0xffffffff >> (32 - TXX9_TIMER_BITS), cd);
        cd->min_delta_ns = clockevent_delta2ns(0xf, cd);
        cd->irq = irq;
+        cd->cpumask = cpumask_of(0),
        clockevents_register_device(cd);
        setup_irq(irq, &txx9tmr_irq);
        printk(KERN_INFO "TXx9: clockevent device at 0x%lx, irq %d\n",
diff --git a/arch/mips/kernel/i8253.c b/arch/mips/kernel/i8253.c
index b6ac55162b9a..f4d187825f96 100644
--- a/arch/mips/kernel/i8253.c
+++ b/arch/mips/kernel/i8253.c
@@ -115,7 +115,7 @@ void __init setup_pit_timer(void)
         * Start pit with the boot cpu mask and make it global after the
         * IO_APIC has been initialized.
         */
-        cd->cpumask = cpumask_of_cpu(cpu);
+        cd->cpumask = cpumask_of(cpu);
        clockevent_set_clock(cd, CLOCK_TICK_RATE);
        cd->max_delta_ns = clockevent_delta2ns(0x7FFF, cd);
        cd->min_delta_ns = clockevent_delta2ns(0xF, cd);
diff --git a/arch/mips/kernel/irq-gic.c b/arch/mips/kernel/irq-gic.c
index f0a4bb19e096..494a49a317e9 100644
--- a/arch/mips/kernel/irq-gic.c
+++ b/arch/mips/kernel/irq-gic.c
@@ -155,7 +155,7 @@ static void gic_unmask_irq(unsigned int irq)
 static DEFINE_SPINLOCK(gic_lock);
-static void gic_set_affinity(unsigned int irq, cpumask_t cpumask)
+static void gic_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
        cpumask_t       tmp = CPU_MASK_NONE;
        unsigned long   flags;
@@ -164,7 +164,7 @@ static void gic_set_affinity(unsigned int irq, cpumask_t cpumask)
        pr_debug(KERN_DEBUG "%s called\n", __func__);
        irq -= _irqbase;
-        cpus_and(tmp, cpumask, cpu_online_map);
+        cpumask_and(&tmp, cpumask, cpu_online_mask);
        if (cpus_empty(tmp))
                return;
@@ -187,7 +187,7 @@ static void gic_set_affinity(unsigned int irq, cpumask_t cpumask)
                set_bit(irq, pcpu_masks[first_cpu(tmp)].pcpu_mask);
        }
-        irq_desc[irq].affinity = cpumask;
+        irq_desc[irq].affinity = *cpumask;
        spin_unlock_irqrestore(&gic_lock, flags);
 }
diff --git a/arch/mips/kernel/smp-cmp.c b/arch/mips/kernel/smp-cmp.c
index ca476c4f62a5..f27beca4b26d 100644
--- a/arch/mips/kernel/smp-cmp.c
+++ b/arch/mips/kernel/smp-cmp.c
@@ -51,10 +51,10 @@ static int __init allowcpus(char *str)
        int len;
        cpus_clear(cpu_allow_map);
-        if (cpulist_parse(str, cpu_allow_map) == 0) {
+        if (cpulist_parse(str, &cpu_allow_map) == 0) {
                cpu_set(0, cpu_allow_map);
                cpus_and(cpu_possible_map, cpu_possible_map, cpu_allow_map);
-                len = cpulist_scnprintf(buf, sizeof(buf)-1, cpu_possible_map);
+                len = cpulist_scnprintf(buf, sizeof(buf)-1, &cpu_possible_map);
                buf[len] = '\0';
                pr_debug("Allowable CPUs: %s\n", buf);
                return 1;
@@ -226,7 +226,7 @@ void __init cmp_smp_setup(void)
        for (i = 1; i < NR_CPUS; i++) {
                if (amon_cpu_avail(i)) {
-                        cpu_set(i, phys_cpu_present_map);
+                        cpu_set(i, cpu_possible_map);
                        __cpu_number_map[i]     = ++ncpu;
                        __cpu_logical_map[ncpu] = i;
                }
diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c
index 87a1816c1f45..6f7ee5ac46ee 100644
--- a/arch/mips/kernel/smp-mt.c
+++ b/arch/mips/kernel/smp-mt.c
@@ -70,7 +70,7 @@ static unsigned int __init smvp_vpe_init(unsigned int tc, unsigned int mvpconf0,
                write_vpe_c0_vpeconf0(tmp);
                /* Record this as available CPU */
-                cpu_set(tc, phys_cpu_present_map);
+                cpu_set(tc, cpu_possible_map);
                __cpu_number_map[tc]    = ++ncpu;
                __cpu_logical_map[ncpu] = tc;
        }
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 8bf88faf5afd..3da94704f816 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -44,15 +44,10 @@
 #include <asm/mipsmtregs.h>
 #endif /* CONFIG_MIPS_MT_SMTC */
-cpumask_t phys_cpu_present_map;         /* Bitmask of available CPUs */
 volatile cpumask_t cpu_callin_map;      /* Bitmask of started secondaries */
-cpumask_t cpu_online_map;               /* Bitmask of currently online CPUs */
 int __cpu_number_map[NR_CPUS];          /* Map physical to logical */
 int __cpu_logical_map[NR_CPUS];         /* Map logical to physical */
-EXPORT_SYMBOL(phys_cpu_present_map);
-EXPORT_SYMBOL(cpu_online_map);
 extern void cpu_idle(void);
 /* Number of TCs (or siblings in Intel speak) per CPU core */
@@ -195,7 +190,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 /* preload SMP state for boot cpu */
 void __devinit smp_prepare_boot_cpu(void)
 {
-        cpu_set(0, phys_cpu_present_map);
+        cpu_set(0, cpu_possible_map);
        cpu_set(0, cpu_online_map);
        cpu_set(0, cpu_callin_map);
 }
diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c
index 897fb2b4751c..b6cca01ff82b 100644
--- a/arch/mips/kernel/smtc.c
+++ b/arch/mips/kernel/smtc.c
@@ -290,7 +290,7 @@ static void smtc_configure_tlb(void)
 * possibly leave some TCs/VPEs as "slave" processors.
 *
 * Use c0_MVPConf0 to find out how many TCs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
+ * cpu_possible_map and the logical/physical mappings.
 */
 int __init smtc_build_cpu_map(int start_cpu_slot)
@@ -304,7 +304,7 @@ int __init smtc_build_cpu_map(int start_cpu_slot)
         */
        ntcs = ((read_c0_mvpconf0() & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT) + 1;
        for (i=start_cpu_slot; i<NR_CPUS && i<ntcs; i++) {
-                cpu_set(i, phys_cpu_present_map);
+                cpu_set(i, cpu_possible_map);
                __cpu_number_map[i] = i;
                __cpu_logical_map[i] = i;
        }
@@ -521,7 +521,7 @@ void smtc_prepare_cpus(int cpus)
         * Pull any physically present but unused TCs out of circulation.
         */
        while (tc < (((val & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT) + 1)) {
-                cpu_clear(tc, phys_cpu_present_map);
+                cpu_clear(tc, cpu_possible_map);
                cpu_clear(tc, cpu_present_map);
                tc++;
        }
diff --git a/arch/mips/mti-malta/malta-smtc.c b/arch/mips/mti-malta/malta-smtc.c
index f84a46a8ae6e..aabd7274507b 100644
--- a/arch/mips/mti-malta/malta-smtc.c
+++ b/arch/mips/mti-malta/malta-smtc.c
@@ -114,9 +114,9 @@ struct plat_smp_ops msmtc_smp_ops = {
 */
-void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity)
+void plat_set_irq_affinity(unsigned int irq, const struct cpumask *affinity)
 {
-        cpumask_t tmask = affinity;
+        cpumask_t tmask = *affinity;
        int cpu = 0;
        void smtc_set_irq_affinity(unsigned int irq, cpumask_t aff);
@@ -139,7 +139,7 @@ void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity)
         * be made to forward to an offline "CPU".
         */
-        for_each_cpu_mask(cpu, affinity) {
+        for_each_cpu(cpu, affinity) {
                if ((cpu_data[cpu].vpe_id != 0) || !cpu_online(cpu))
                        cpu_clear(cpu, tmask);
        }
diff --git a/arch/mips/nxp/pnx8550/common/time.c b/arch/mips/nxp/pnx8550/common/time.c
index 62f495b57f93..cf293b279098 100644
--- a/arch/mips/nxp/pnx8550/common/time.c
+++ b/arch/mips/nxp/pnx8550/common/time.c
@@ -102,6 +102,7 @@ __init void plat_time_init(void)
        unsigned int p;
        unsigned int pow2p;
+        pnx8xxx_clockevent.cpumask = cpu_none_mask;
        clockevents_register_device(&pnx8xxx_clockevent);
        clocksource_register(&pnx_clocksource);
diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c
index 3a7df647ca77..f78c29b68d77 100644
--- a/arch/mips/pmc-sierra/yosemite/smp.c
+++ b/arch/mips/pmc-sierra/yosemite/smp.c
@@ -141,7 +141,7 @@ static void __cpuinit yos_boot_secondary(int cpu, struct task_struct *idle)
 }
 /*
- * Detect available CPUs, populate phys_cpu_present_map before smp_init
+ * Detect available CPUs, populate cpu_possible_map before smp_init
 *
 * We don't want to start the secondary CPU yet nor do we have a nice probing
 * feature in PMON so we just assume presence of the secondary core.
@@ -150,10 +150,10 @@ static void __init yos_smp_setup(void)
 {
        int i;
-        cpus_clear(phys_cpu_present_map);
+        cpus_clear(cpu_possible_map);
        for (i = 0; i < 2; i++) {
-                cpu_set(i, phys_cpu_present_map);
+                cpu_set(i, cpu_possible_map);
                __cpu_number_map[i]     = i;
                __cpu_logical_map[i]    = i;
        }
diff --git a/arch/mips/sgi-ip27/ip27-smp.c b/arch/mips/sgi-ip27/ip27-smp.c
index ba5cdebeaf0d..5b47d6b65275 100644
--- a/arch/mips/sgi-ip27/ip27-smp.c
+++ b/arch/mips/sgi-ip27/ip27-smp.c
@@ -76,7 +76,7 @@ static int do_cpumask(cnodeid_t cnode, nasid_t nasid, int highest)
                        /* Only let it join in if it's marked enabled */
                        if ((acpu->cpu_info.flags & KLINFO_ENABLE) &&
                            (tot_cpus_found != NR_CPUS)) {
-                                cpu_set(cpuid, phys_cpu_present_map);
+                                cpu_set(cpuid, cpu_possible_map);
                                alloc_cpupda(cpuid, tot_cpus_found);
                                cpus_found++;
                                tot_cpus_found++;
diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c
index 1327c2746fb7..f024057a35f8 100644
--- a/arch/mips/sgi-ip27/ip27-timer.c
+++ b/arch/mips/sgi-ip27/ip27-timer.c
@@ -134,7 +134,7 @@ void __cpuinit hub_rt_clock_event_init(void)
        cd->min_delta_ns        = clockevent_delta2ns(0x300, cd);
        cd->rating              = 200;
        cd->irq                 = irq;
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        cd->set_next_event      = rt_next_event;
        cd->set_mode            = rt_set_mode;
        clockevents_register_device(cd);
diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c
index a35818ed4263..12b465d404df 100644
--- a/arch/mips/sibyte/bcm1480/irq.c
+++ b/arch/mips/sibyte/bcm1480/irq.c
@@ -50,7 +50,7 @@ static void enable_bcm1480_irq(unsigned int irq);
 static void disable_bcm1480_irq(unsigned int irq);
 static void ack_bcm1480_irq(unsigned int irq);
 #ifdef CONFIG_SMP
-static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask);
+static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask);
 #endif
 #ifdef CONFIG_PCI
@@ -109,7 +109,7 @@ void bcm1480_unmask_irq(int cpu, int irq)
 }
 #ifdef CONFIG_SMP
-static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask)
+static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
        int i = 0, old_cpu, cpu, int_on, k;
        u64 cur_ints;
@@ -117,11 +117,11 @@ static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask)
        unsigned long flags;
        unsigned int irq_dirty;
-        if (cpus_weight(mask) != 1) {
+        if (cpumask_weight(mask) != 1) {
                printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq);
                return;
        }
-        i = first_cpu(mask);
+        i = cpumask_first(mask);
        /* Convert logical CPU to physical CPU */
        cpu = cpu_logical_map(i);
diff --git a/arch/mips/sibyte/bcm1480/smp.c b/arch/mips/sibyte/bcm1480/smp.c
index bd9eeb43ed0e..dddfda8e8294 100644
--- a/arch/mips/sibyte/bcm1480/smp.c
+++ b/arch/mips/sibyte/bcm1480/smp.c
@@ -136,7 +136,7 @@ static void __cpuinit bcm1480_boot_secondary(int cpu, struct task_struct *idle)
 /*
 * Use CFE to find out how many CPUs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
+ * cpu_possible_map and the logical/physical mappings.
 * XXXKW will the boot CPU ever not be physical 0?
 *
 * Common setup before any secondaries are started
@@ -145,14 +145,14 @@ static void __init bcm1480_smp_setup(void)
 {
        int i, num;
-        cpus_clear(phys_cpu_present_map);
+        cpus_clear(cpu_possible_map);
-        cpu_set(0, phys_cpu_present_map);
+        cpu_set(0, cpu_possible_map);
        __cpu_number_map[0] = 0;
        __cpu_logical_map[0] = 0;
        for (i = 1, num = 0; i < NR_CPUS; i++) {
                if (cfe_cpu_stop(i) == 0) {
-                        cpu_set(i, phys_cpu_present_map);
+                        cpu_set(i, cpu_possible_map);
                        __cpu_number_map[i] = ++num;
                        __cpu_logical_map[num] = i;
                }
diff --git a/arch/mips/sibyte/sb1250/irq.c b/arch/mips/sibyte/sb1250/irq.c
index a5158483986e..808ac2959b8c 100644
--- a/arch/mips/sibyte/sb1250/irq.c
+++ b/arch/mips/sibyte/sb1250/irq.c
@@ -50,7 +50,7 @@ static void enable_sb1250_irq(unsigned int irq);
 static void disable_sb1250_irq(unsigned int irq);
 static void ack_sb1250_irq(unsigned int irq);
 #ifdef CONFIG_SMP
-static void sb1250_set_affinity(unsigned int irq, cpumask_t mask);
+static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask);
 #endif
 #ifdef CONFIG_SIBYTE_HAS_LDT
@@ -103,16 +103,16 @@ void sb1250_unmask_irq(int cpu, int irq)
 }
 #ifdef CONFIG_SMP
-static void sb1250_set_affinity(unsigned int irq, cpumask_t mask)
+static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
        int i = 0, old_cpu, cpu, int_on;
        u64 cur_ints;
        struct irq_desc *desc = irq_desc + irq;
        unsigned long flags;
-        i = first_cpu(mask);
+        i = cpumask_first(mask);
-        if (cpus_weight(mask) > 1) {
+        if (cpumask_weight(mask) > 1) {
                printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq);
                return;
        }
diff --git a/arch/mips/sibyte/sb1250/smp.c b/arch/mips/sibyte/sb1250/smp.c
index 0734b933e969..5950a288a7da 100644
--- a/arch/mips/sibyte/sb1250/smp.c
+++ b/arch/mips/sibyte/sb1250/smp.c
@@ -124,7 +124,7 @@ static void __cpuinit sb1250_boot_secondary(int cpu, struct task_struct *idle)
 /*
 * Use CFE to find out how many CPUs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
+ * cpu_possible_map and the logical/physical mappings.
 * XXXKW will the boot CPU ever not be physical 0?
 *
 * Common setup before any secondaries are started
@@ -133,14 +133,14 @@ static void __init sb1250_smp_setup(void)
 {
        int i, num;
-        cpus_clear(phys_cpu_present_map);
+        cpus_clear(cpu_possible_map);
-        cpu_set(0, phys_cpu_present_map);
+        cpu_set(0, cpu_possible_map);
        __cpu_number_map[0] = 0;
        __cpu_logical_map[0] = 0;
        for (i = 1, num = 0; i < NR_CPUS; i++) {
                if (cfe_cpu_stop(i) == 0) {
-                        cpu_set(i, phys_cpu_present_map);
+                        cpu_set(i, cpu_possible_map);
                        __cpu_number_map[i] = ++num;
                        __cpu_logical_map[num] = i;
                }
diff --git a/arch/mips/sni/time.c b/arch/mips/sni/time.c
index 796e3ce28720..69f5f88711cc 100644
--- a/arch/mips/sni/time.c
+++ b/arch/mips/sni/time.c
@@ -80,7 +80,7 @@ static void __init sni_a20r_timer_setup(void)
        struct irqaction *action = &a20r_irqaction;
        unsigned int cpu = smp_processor_id();
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        clockevents_register_device(cd);
        action->dev_id = cd;
        setup_irq(SNI_A20R_IRQ_TIMER, &a20r_irqaction);
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 644a70b1b04e..aacf11d33723 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -11,6 +11,7 @@ config PARISC
        select HAVE_OPROFILE
        select RTC_CLASS
        select RTC_DRV_PARISC
+        select INIT_ALL_POSSIBLE
        help
          The PA-RISC microprocessor is designed by Hewlett-Packard and used
          in many of their workstations & servers (HP9000 700 and 800 series,
diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c
index 23ef950df008..4cea935e2f99 100644
--- a/arch/parisc/kernel/irq.c
+++ b/arch/parisc/kernel/irq.c
@@ -131,12 +131,12 @@ int cpu_check_affinity(unsigned int irq, cpumask_t *dest)
        return 0;
 }
-static void cpu_set_affinity_irq(unsigned int irq, cpumask_t dest)
+static void cpu_set_affinity_irq(unsigned int irq, const struct cpumask *dest)
 {
-        if (cpu_check_affinity(irq, &dest))
+        if (cpu_check_affinity(irq, dest))
                return;
-        irq_desc[irq].affinity = dest;
+        irq_desc[irq].affinity = *dest;
 }
 #endif
diff --git a/arch/parisc/kernel/smp.c b/arch/parisc/kernel/smp.c
index d47f3975c9c6..80bc000523fa 100644
--- a/arch/parisc/kernel/smp.c
+++ b/arch/parisc/kernel/smp.c
@@ -67,21 +67,6 @@ static volatile int cpu_now_booting __read_mostly = 0;	/* track which CPU is boo
 static int parisc_max_cpus __read_mostly = 1;
-/* online cpus are ones that we've managed to bring up completely
- * possible cpus are all valid cpu 
- * present cpus are all detected cpu
- *
- * On startup we bring up the "possible" cpus. Since we discover
- * CPUs later, we add them as hotplug, so the possible cpu mask is
- * empty in the beginning.
- */
-cpumask_t cpu_online_map   __read_mostly = CPU_MASK_NONE;       /* Bitmap of online CPUs */
-cpumask_t cpu_possible_map __read_mostly = CPU_MASK_ALL;        /* Bitmap of Present CPUs */
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(cpu_possible_map);
 DEFINE_PER_CPU(spinlock_t, ipi_lock) = SPIN_LOCK_UNLOCKED;
 enum ipi_message_type {
diff --git a/arch/powerpc/include/asm/topology.h b/arch/powerpc/include/asm/topology.h
index c32da6f97999..bcf25c2b8d21 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;
+        return cpumask_first(cpumask_of_node(node));
-        tmp = node_to_cpumask(node);
-        return first_cpu(tmp);
 }
 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) {            \
        .span                   = CPU_MASK_NONE,        \
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index ac222d0ab12e..23b8b5e36f98 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -237,7 +237,7 @@ void fixup_irqs(cpumask_t map)
                        mask = map;
                }
                if (irq_desc[irq].chip->set_affinity)
-                        irq_desc[irq].chip->set_affinity(irq, mask);
+                        irq_desc[irq].chip->set_affinity(irq, &mask);
                else if (irq_desc[irq].action && !(warned++))
                        printk("Cannot set affinity for irq %i\n", irq);
        }
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 8ac3f721d235..65484b2200b3 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -59,13 +59,9 @@
 struct thread_info *secondary_ti;
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-cpumask_t cpu_online_map = CPU_MASK_NONE;
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map) = CPU_MASK_NONE;
 DEFINE_PER_CPU(cpumask_t, cpu_core_map) = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(cpu_possible_map);
 EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
 EXPORT_PER_CPU_SYMBOL(cpu_core_map);
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index e1f3a5140429..99f1ddd68582 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -844,7 +844,7 @@ static void register_decrementer_clockevent(int cpu)
        struct clock_event_device *dec = &per_cpu(decrementers, cpu).event;
        *dec = decrementer_clockevent;
-        dec->cpumask = cpumask_of_cpu(cpu);
+        dec->cpumask = cpumask_of(cpu);
        printk(KERN_DEBUG "clockevent: %s mult[%lx] shift[%d] cpu[%d]\n",
               dec->name, dec->mult, dec->shift, cpu);
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);
+                const struct cpumask *spumask = cpumask_of_node(spu->node),
-                cpumask_t cpumask = node_to_cpumask(cpu_to_node(cpu));
+                        *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/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index f7a69021b7bf..84e058f1e1cc 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -332,7 +332,7 @@ static void xics_eoi_lpar(unsigned int virq)
        lpar_xirr_info_set((0xff << 24) | irq);
 }
-static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
+static void xics_set_affinity(unsigned int virq, const struct cpumask *cpumask)
 {
        unsigned int irq;
        int status;
@@ -870,7 +870,7 @@ void xics_migrate_irqs_away(void)
                /* Reset affinity to all cpus */
                irq_desc[virq].affinity = CPU_MASK_ALL;
-                desc->chip->set_affinity(virq, CPU_MASK_ALL);
+                desc->chip->set_affinity(virq, cpu_all_mask);
 unlock:
                spin_unlock_irqrestore(&desc->lock, flags);
        }
diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index c82babb70074..3e0d89dcdba2 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -806,7 +806,7 @@ static void mpic_end_ipi(unsigned int irq)
 #endif /* CONFIG_SMP */
-void mpic_set_affinity(unsigned int irq, cpumask_t cpumask)
+void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
        struct mpic *mpic = mpic_from_irq(irq);
        unsigned int src = mpic_irq_to_hw(irq);
@@ -818,7 +818,7 @@ void mpic_set_affinity(unsigned int irq, cpumask_t cpumask)
        } else {
                cpumask_t tmp;
-                cpus_and(tmp, cpumask, cpu_online_map);
+                cpumask_and(&tmp, cpumask, cpu_online_mask);
                mpic_irq_write(src, MPIC_INFO(IRQ_DESTINATION),
                               mpic_physmask(cpus_addr(tmp)[0]));
diff --git a/arch/powerpc/sysdev/mpic.h b/arch/powerpc/sysdev/mpic.h
index 6209c62a426d..3cef2af10f42 100644
--- a/arch/powerpc/sysdev/mpic.h
+++ b/arch/powerpc/sysdev/mpic.h
@@ -36,6 +36,6 @@ static inline int mpic_pasemi_msi_init(struct mpic *mpic)
 extern int mpic_set_irq_type(unsigned int virq, unsigned int flow_type);
 extern void mpic_set_vector(unsigned int virq, unsigned int vector);
-extern void mpic_set_affinity(unsigned int irq, cpumask_t cpumask);
+extern void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask);
 #endif /* _POWERPC_SYSDEV_MPIC_H */
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 8152fefc97b9..19577aeffd7b 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -83,6 +83,7 @@ config S390
        select HAVE_KRETPROBES
        select HAVE_KVM if 64BIT
        select HAVE_ARCH_TRACEHOOK
+        select INIT_ALL_POSSIBLE
 source "init/Kconfig"
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/smp.c b/arch/s390/kernel/smp.c
index 6fc78541dc57..3ed5c7a83c6c 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -55,12 +55,6 @@
 struct _lowcore *lowcore_ptr[NR_CPUS];
 EXPORT_SYMBOL(lowcore_ptr);
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
-cpumask_t cpu_possible_map = CPU_MASK_ALL;
-EXPORT_SYMBOL(cpu_possible_map);
 static struct task_struct *current_set[NR_CPUS];
 static u8 smp_cpu_type;
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 5be981a36c3e..d649600df5b9 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -160,7 +160,7 @@ void init_cpu_timer(void)
        cd->min_delta_ns        = 1;
        cd->max_delta_ns        = LONG_MAX;
        cd->rating              = 400;
-        cd->cpumask             = cpumask_of_cpu(cpu);
+        cd->cpumask             = cpumask_of(cpu);
        cd->set_next_event      = s390_next_event;
        cd->set_mode            = s390_set_mode;
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/smp.h b/arch/sh/include/asm/smp.h
index 85b660c17eb0..c24e9c6a1736 100644
--- a/arch/sh/include/asm/smp.h
+++ b/arch/sh/include/asm/smp.h
@@ -31,7 +31,7 @@ enum {
 };
 void smp_message_recv(unsigned int msg);
-void smp_timer_broadcast(cpumask_t mask);
+void smp_timer_broadcast(const struct cpumask *mask);
 void local_timer_interrupt(void);
 void local_timer_setup(unsigned int cpu);
diff --git a/arch/sh/include/asm/topology.h b/arch/sh/include/asm/topology.h
index 95f0085e098a..9aa160d0efe5 100644
--- a/arch/sh/include/asm/topology.h
+++ b/arch/sh/include/asm/topology.h
@@ -33,6 +33,7 @@
 #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/sh/kernel/smp.c b/arch/sh/kernel/smp.c
index 3c5ad1660bbc..8f4027412614 100644
--- a/arch/sh/kernel/smp.c
+++ b/arch/sh/kernel/smp.c
@@ -31,12 +31,6 @@
 int __cpu_number_map[NR_CPUS];          /* Map physical to logical */
 int __cpu_logical_map[NR_CPUS];         /* Map logical to physical */
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
 static inline void __init smp_store_cpu_info(unsigned int cpu)
 {
        struct sh_cpuinfo *c = cpu_data + cpu;
@@ -190,11 +184,11 @@ void arch_send_call_function_single_ipi(int cpu)
        plat_send_ipi(cpu, SMP_MSG_FUNCTION_SINGLE);
 }
-void smp_timer_broadcast(cpumask_t mask)
+void smp_timer_broadcast(const struct cpumask *mask)
 {
        int cpu;
-        for_each_cpu_mask(cpu, mask)
+        for_each_cpu(cpu, mask)
                plat_send_ipi(cpu, SMP_MSG_TIMER);
 }
diff --git a/arch/sh/kernel/timers/timer-broadcast.c b/arch/sh/kernel/timers/timer-broadcast.c
index c2317635230f..96e8eaea1e62 100644
--- a/arch/sh/kernel/timers/timer-broadcast.c
+++ b/arch/sh/kernel/timers/timer-broadcast.c
@@ -51,7 +51,7 @@ void __cpuinit local_timer_setup(unsigned int cpu)
        clk->mult               = 1;
        clk->set_mode           = dummy_timer_set_mode;
        clk->broadcast          = smp_timer_broadcast;
-        clk->cpumask            = cpumask_of_cpu(cpu);
+        clk->cpumask            = cpumask_of(cpu);
        clockevents_register_device(clk);
 }
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index 3c61ddd4d43e..0db3f9510336 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -263,7 +263,7 @@ static int tmu_timer_init(void)
        tmu0_clockevent.min_delta_ns =
                        clockevent_delta2ns(1, &tmu0_clockevent);
-        tmu0_clockevent.cpumask = cpumask_of_cpu(0);
+        tmu0_clockevent.cpumask = cpumask_of(0);
        clockevents_register_device(&tmu0_clockevent);
diff --git a/arch/sparc/include/asm/smp_32.h b/arch/sparc/include/asm/smp_32.h
index a8180e546a48..8408d9d2a662 100644
--- a/arch/sparc/include/asm/smp_32.h
+++ b/arch/sparc/include/asm/smp_32.h
@@ -29,8 +29,6 @@
 */
 extern unsigned char boot_cpu_id;
-extern cpumask_t phys_cpu_present_map;
-#define cpu_possible_map phys_cpu_present_map
 typedef void (*smpfunc_t)(unsigned long, unsigned long, unsigned long,
                       unsigned long, unsigned long);
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;
+        return cpumask_first(cpumask_of_node(node));
-        tmp = node_to_cpumask(node);
-        return first_cpu(tmp);
 }
 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/irq_64.c b/arch/sparc/kernel/irq_64.c
index a3ea2bcb95de..cab8e0286871 100644
--- a/arch/sparc/kernel/irq_64.c
+++ b/arch/sparc/kernel/irq_64.c
@@ -312,7 +312,8 @@ static void sun4u_irq_enable(unsigned int virt_irq)
        }
 }
-static void sun4u_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4u_set_affinity(unsigned int virt_irq,
+                               const struct cpumask *mask)
 {
        sun4u_irq_enable(virt_irq);
 }
@@ -362,7 +363,8 @@ static void sun4v_irq_enable(unsigned int virt_irq)
                       ino, err);
 }
-static void sun4v_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4v_set_affinity(unsigned int virt_irq,
+                               const struct cpumask *mask)
 {
        unsigned int ino = virt_irq_table[virt_irq].dev_ino;
        unsigned long cpuid = irq_choose_cpu(virt_irq);
@@ -429,7 +431,8 @@ static void sun4v_virq_enable(unsigned int virt_irq)
                       dev_handle, dev_ino, err);
 }
-static void sun4v_virt_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4v_virt_set_affinity(unsigned int virt_irq,
+                                    const struct cpumask *mask)
 {
        unsigned long cpuid, dev_handle, dev_ino;
        int err;
@@ -851,7 +854,7 @@ void fixup_irqs(void)
                    !(irq_desc[irq].status & IRQ_PER_CPU)) {
                        if (irq_desc[irq].chip->set_affinity)
                                irq_desc[irq].chip->set_affinity(irq,
-                                        irq_desc[irq].affinity);
+                                        &irq_desc[irq].affinity);
                }
                spin_unlock_irqrestore(&irq_desc[irq].lock, flags);
        }
diff --git a/arch/sparc/kernel/of_device_64.c b/arch/sparc/kernel/of_device_64.c
index 46e231f7c5ce..4873f28905b0 100644
--- a/arch/sparc/kernel/of_device_64.c
+++ b/arch/sparc/kernel/of_device_64.c
@@ -778,9 +778,9 @@ 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);
+                irq_set_affinity(irq, &numa_mask);
        }
        return irq;
diff --git a/arch/sparc/kernel/pci_msi.c b/arch/sparc/kernel/pci_msi.c
index 2e680f34f727..4ef282e81912 100644
--- a/arch/sparc/kernel/pci_msi.c
+++ b/arch/sparc/kernel/pci_msi.c
@@ -286,9 +286,9 @@ 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);
+                irq_set_affinity(irq, &numa_mask);
        }
        err = request_irq(irq, sparc64_msiq_interrupt, 0,
                          "MSIQ",
diff --git a/arch/sparc/kernel/smp_32.c b/arch/sparc/kernel/smp_32.c
index e396c1f17a92..1e5ac4e282e1 100644
--- a/arch/sparc/kernel/smp_32.c
+++ b/arch/sparc/kernel/smp_32.c
@@ -39,8 +39,6 @@ volatile unsigned long cpu_callin_map[NR_CPUS] __cpuinitdata = {0,};
 unsigned char boot_cpu_id = 0;
 unsigned char boot_cpu_id4 = 0; /* boot_cpu_id << 2 */
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-cpumask_t phys_cpu_present_map = CPU_MASK_NONE;
 cpumask_t smp_commenced_mask = CPU_MASK_NONE;
 /* The only guaranteed locking primitive available on all Sparc
@@ -334,7 +332,7 @@ void __init smp_setup_cpu_possible_map(void)
        instance = 0;
        while (!cpu_find_by_instance(instance, NULL, &mid)) {
                if (mid < NR_CPUS) {
-                        cpu_set(mid, phys_cpu_present_map);
+                        cpu_set(mid, cpu_possible_map);
                        cpu_set(mid, cpu_present_map);
                }
                instance++;
@@ -354,7 +352,7 @@ void __init smp_prepare_boot_cpu(void)
        current_thread_info()->cpu = cpuid;
        cpu_set(cpuid, cpu_online_map);
-        cpu_set(cpuid, phys_cpu_present_map);
+        cpu_set(cpuid, cpu_possible_map);
 }
 int __cpuinit __cpu_up(unsigned int cpu)
diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c
index bfe99d82d458..46329799f346 100644
--- a/arch/sparc/kernel/smp_64.c
+++ b/arch/sparc/kernel/smp_64.c
@@ -49,14 +49,10 @@
 int sparc64_multi_core __read_mostly;
-cpumask_t cpu_possible_map __read_mostly = CPU_MASK_NONE;
-cpumask_t cpu_online_map __read_mostly = CPU_MASK_NONE;
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map) = CPU_MASK_NONE;
 cpumask_t cpu_core_map[NR_CPUS] __read_mostly =
        { [0 ... NR_CPUS-1] = CPU_MASK_NONE };
-EXPORT_SYMBOL(cpu_possible_map);
-EXPORT_SYMBOL(cpu_online_map);
 EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
 EXPORT_SYMBOL(cpu_core_map);
diff --git a/arch/sparc/kernel/sparc_ksyms_32.c b/arch/sparc/kernel/sparc_ksyms_32.c
index a4d45fc29b21..e1e97639231b 100644
--- a/arch/sparc/kernel/sparc_ksyms_32.c
+++ b/arch/sparc/kernel/sparc_ksyms_32.c
@@ -112,10 +112,6 @@ EXPORT_PER_CPU_SYMBOL(__cpu_data);
 #ifdef CONFIG_SMP
 /* IRQ implementation. */
 EXPORT_SYMBOL(synchronize_irq);
-/* CPU online map and active count. */
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(phys_cpu_present_map);
 #endif
 EXPORT_SYMBOL(__udelay);
diff --git a/arch/sparc/kernel/time_64.c b/arch/sparc/kernel/time_64.c
index 141da3759091..9df8f095a8b1 100644
--- a/arch/sparc/kernel/time_64.c
+++ b/arch/sparc/kernel/time_64.c
@@ -763,7 +763,7 @@ void __devinit setup_sparc64_timer(void)
        sevt = &__get_cpu_var(sparc64_events);
        memcpy(sevt, &sparc64_clockevent, sizeof(*sevt));
-        sevt->cpumask = cpumask_of_cpu(smp_processor_id());
+        sevt->cpumask = cpumask_of(smp_processor_id());
        clockevents_register_device(sevt);
 }
diff --git a/arch/um/kernel/smp.c b/arch/um/kernel/smp.c
index 045772142844..98351c78bc81 100644
--- a/arch/um/kernel/smp.c
+++ b/arch/um/kernel/smp.c
@@ -25,13 +25,6 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 #include "irq_user.h"
 #include "os.h"
-/* CPU online map, set by smp_boot_cpus */
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(cpu_possible_map);
 /* Per CPU bogomips and other parameters
 * The only piece used here is the ipi pipe, which is set before SMP is
 * started and never changed.
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index 47f04f4a3464..b13a87a3ec95 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -50,7 +50,7 @@ static int itimer_next_event(unsigned long delta,
 static struct clock_event_device itimer_clockevent = {
        .name           = "itimer",
        .rating         = 250,
-        .cpumask        = CPU_MASK_ALL,
+        .cpumask        = cpu_all_mask,
        .features       = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
        .set_mode       = itimer_set_mode,
        .set_next_event = itimer_next_event,
diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h
index 66834c41c049..a977de23cb4d 100644
--- a/arch/x86/include/asm/pci.h
+++ b/arch/x86/include/asm/pci.h
@@ -102,9 +102,9 @@ extern void pci_iommu_alloc(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;
 }
@@ -113,6 +113,12 @@ 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 ff386ff50ed7..168203c0c316 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 cpumask_first(cpumask_of_node(node));
-        return first_cpu(*mask);
 }
 #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/apic.c b/arch/x86/kernel/apic.c
index b5229affb953..6107b41da9a5 100644
--- a/arch/x86/kernel/apic.c
+++ b/arch/x86/kernel/apic.c
@@ -142,7 +142,7 @@ 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);
 /*
@@ -455,10 +455,10 @@ 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);
+        send_IPI_mask(*mask, LOCAL_TIMER_VECTOR);
 #endif
 }
@@ -471,7 +471,7 @@ 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);
 }
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 68b5d8681cbb..15cf14e9bf26 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -626,8 +626,8 @@ 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):
+                        cpulist_scnprintf(buf, len-2, mask) :
-                        cpumask_scnprintf(buf, len-2, *mask);
+                        cpumask_scnprintf(buf, len-2, mask);
                buf[n++] = '\n';
                buf[n] = '\0';
        }
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 f6ea94b74da1..e7745961ed31 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -550,13 +550,14 @@ static void set_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mask)
        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;
        desc = irq_to_desc(irq);
-        set_ioapic_affinity_irq_desc(desc, mask);
+        set_ioapic_affinity_irq_desc(desc, *mask);
 }
 #endif /* CONFIG_SMP */
@@ -2392,7 +2393,7 @@ 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);
                }
        }
@@ -2412,11 +2413,12 @@ static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mas
        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);
-        set_ir_ioapic_affinity_irq_desc(desc, mask);
+        set_ir_ioapic_affinity_irq_desc(desc, *mask);
 }
 #endif
@@ -3285,7 +3287,7 @@ 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;
@@ -3293,17 +3295,16 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
        unsigned int dest;
        cpumask_t tmp;
-        cpus_and(tmp, mask, cpu_online_map);
+        if (!cpumask_intersects(mask, cpu_online_mask))
-        if (cpus_empty(tmp))
                return;
        cfg = desc->chip_data;
-        if (assign_irq_vector(irq, cfg, mask))
+        if (assign_irq_vector(irq, cfg, *mask))
                return;
-        set_extra_move_desc(desc, mask);
+        set_extra_move_desc(desc, *mask);
-        cpus_and(tmp, cfg->domain, mask);
+        cpumask_and(&tmp, &cfg->domain, mask);
        dest = cpu_mask_to_apicid(tmp);
        read_msi_msg_desc(desc, &msg);
@@ -3314,14 +3315,15 @@ 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;
+        cpumask_copy(&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;
@@ -3329,20 +3331,19 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
        cpumask_t tmp, cleanup_mask;
        struct irte irte;
-        cpus_and(tmp, mask, cpu_online_map);
+        if (!cpumask_intersects(mask, cpu_online_mask))
-        if (cpus_empty(tmp))
                return;
        if (get_irte(irq, &irte))
                return;
        cfg = desc->chip_data;
-        if (assign_irq_vector(irq, cfg, mask))
+        if (assign_irq_vector(irq, cfg, *mask))
                return;
-        set_extra_move_desc(desc, mask);
+        set_extra_move_desc(desc, *mask);
-        cpus_and(tmp, cfg->domain, mask);
+        cpumask_and(&tmp, &cfg->domain, mask);
        dest = cpu_mask_to_apicid(tmp);
        irte.vector = cfg->vector;
@@ -3365,7 +3366,7 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
                cfg->move_in_progress = 0;
        }
-        desc->affinity = mask;
+        cpumask_copy(&desc->affinity, mask);
 }
 #endif
@@ -3556,7 +3557,7 @@ 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;
@@ -3564,17 +3565,16 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
        unsigned int dest;
        cpumask_t tmp;
-        cpus_and(tmp, mask, cpu_online_map);
+        if (!cpumask_intersects(mask, cpu_online_mask))
-        if (cpus_empty(tmp))
                return;
        cfg = desc->chip_data;
-        if (assign_irq_vector(irq, cfg, mask))
+        if (assign_irq_vector(irq, cfg, *mask))
                return;
-        set_extra_move_desc(desc, mask);
+        set_extra_move_desc(desc, *mask);
-        cpus_and(tmp, cfg->domain, mask);
+        cpumask_and(&tmp, &cfg->domain, mask);
        dest = cpu_mask_to_apicid(tmp);
        dmar_msi_read(irq, &msg);
@@ -3585,7 +3585,7 @@ 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;
+        cpumask_copy(&desc->affinity, mask);
 }
 #endif /* CONFIG_SMP */
@@ -3619,7 +3619,7 @@ 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;
@@ -3627,17 +3627,16 @@ static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
        unsigned int dest;
        cpumask_t tmp;
-        cpus_and(tmp, mask, cpu_online_map);
+        if (!cpumask_intersects(mask, cpu_online_mask))
-        if (cpus_empty(tmp))
                return;
        cfg = desc->chip_data;
-        if (assign_irq_vector(irq, cfg, mask))
+        if (assign_irq_vector(irq, cfg, *mask))
                return;
-        set_extra_move_desc(desc, mask);
+        set_extra_move_desc(desc, *mask);
-        cpus_and(tmp, cfg->domain, mask);
+        cpumask_and(&tmp, &cfg->domain, mask);
        dest = cpu_mask_to_apicid(tmp);
        hpet_msi_read(irq, &msg);
@@ -3648,7 +3647,7 @@ 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;
+        cpumask_copy(&desc->affinity, mask);
 }
 #endif /* CONFIG_SMP */
@@ -3703,28 +3702,27 @@ 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 (!cpumask_intersects(mask, cpu_online_mask))
-        if (cpus_empty(tmp))
                return;
        cfg = desc->chip_data;
-        if (assign_irq_vector(irq, cfg, mask))
+        if (assign_irq_vector(irq, cfg, *mask))
                return;
-        set_extra_move_desc(desc, mask);
+        set_extra_move_desc(desc, *mask);
-        cpus_and(tmp, cfg->domain, mask);
+        cpumask_and(&tmp, &cfg->domain, mask);
        dest = cpu_mask_to_apicid(tmp);
        target_ht_irq(irq, dest, cfg->vector);
-        desc->affinity = mask;
+        cpumask_copy(&desc->affinity, mask);
 }
 #endif
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 119fc9c8ff7f..9cf9cbbf7a02 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -253,7 +253,7 @@ void fixup_irqs(cpumask_t map)
                        mask = map;
                }
                if (desc->chip->set_affinity)
-                        desc->chip->set_affinity(irq, mask);
+                        desc->chip->set_affinity(irq, &mask);
                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..54c69d47a771 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -115,7 +115,7 @@ void fixup_irqs(cpumask_t map)
                        desc->chip->mask(irq);
                if (desc->chip->set_affinity)
-                        desc->chip->set_affinity(irq, mask);
+                        desc->chip->set_affinity(irq, &mask);
                else if (!(warned++))
                        set_affinity = 0;
diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c
index 3b599518c322..c12314c9e86f 100644
--- a/arch/x86/kernel/mfgpt_32.c
+++ b/arch/x86/kernel/mfgpt_32.c
@@ -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/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index ae0c0d3bb770..8e8b1193add5 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -282,7 +282,7 @@ 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 +334,25 @@ 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 f8500c969442..c5392058cd07 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)
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/lguest/boot.c b/arch/x86/lguest/boot.c
index 50a779264bb1..a7ed208f81e3 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -738,7 +738,7 @@ static void lguest_time_init(void)
        /* We can't set cpumask in the initializer: damn C limitations!  Set it
         * here and register our timer device. */
-        lguest_clockevent.cpumask = cpumask_of_cpu(0);
+        lguest_clockevent.cpumask = cpumask_of(0);
        clockevents_register_device(&lguest_clockevent);
        /* Finally, we unblock the timer interrupt. */
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 52145007bd7e..9c990185e9f2 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -63,11 +63,6 @@ static int voyager_extended_cpus = 1;
 /* Used for the invalidate map that's also checked in the spinlock */
 static volatile unsigned long smp_invalidate_needed;
-/* Bitmask of currently online CPUs - used by setup.c for
-   /proc/cpuinfo, visible externally but still physical */
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
 /* Bitmask of CPUs present in the system - exported by i386_syms.c, used
 * by scheduler but indexed physically */
 cpumask_t phys_cpu_present_map = CPU_MASK_NONE;
@@ -218,8 +213,6 @@ static cpumask_t smp_commenced_mask = CPU_MASK_NONE;
 /* This is for the new dynamic CPU boot code */
 cpumask_t cpu_callin_map = CPU_MASK_NONE;
 cpumask_t cpu_callout_map = CPU_MASK_NONE;
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_possible_map);
 /* The per processor IRQ masks (these are usually kept in sync) */
 static __u16 vic_irq_mask[NR_CPUS] __cacheline_aligned;
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index c9f7cda48ed7..65d75a6be0ba 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -437,7 +437,7 @@ void xen_setup_timer(int cpu)
        evt = &per_cpu(xen_clock_events, cpu);
        memcpy(evt, xen_clockevent, sizeof(*evt));
-        evt->cpumask = cpumask_of_cpu(cpu);
+        evt->cpumask = cpumask_of(cpu);
        evt->irq = irq;
        setup_runstate_info(cpu);
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);
+        const struct cpumask *mask = cpu_coregroup_mask(cpu);
-        return first_cpu(mask);
+        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 64f5d54f7edc..2aef96f20b30 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -109,7 +109,7 @@ static SYSDEV_ATTR(crash_notes, 0400, show_crash_notes, NULL);
 */
 static ssize_t print_cpus_map(char *buf, cpumask_t *map)
 {
-        int n = cpulist_scnprintf(buf, PAGE_SIZE-2, *map);
+        int n = cpulist_scnprintf(buf, PAGE_SIZE-2, map);
        buf[n++] = '\n';
        buf[n] = '\0';
@@ -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/base/node.c b/drivers/base/node.c
index f5207090885a..91636cd8b6c9 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -30,8 +30,8 @@ static ssize_t node_read_cpumap(struct sys_device *dev, int type, char *buf)
        BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1));
        len = type?
-                cpulist_scnprintf(buf, PAGE_SIZE-2, *mask):
+                cpulist_scnprintf(buf, PAGE_SIZE-2, mask) :
-                cpumask_scnprintf(buf, PAGE_SIZE-2, *mask);
+                cpumask_scnprintf(buf, PAGE_SIZE-2, mask);
        buf[len++] = '\n';
        buf[len] = '\0';
        return len;
diff --git a/drivers/base/topology.c b/drivers/base/topology.c
index 199cd97e32e6..a8bc1cbcfa7c 100644
--- a/drivers/base/topology.c
+++ b/drivers/base/topology.c
@@ -49,8 +49,8 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf)
        if (len > 1) {
                n = type?
-                        cpulist_scnprintf(buf, len-2, *mask):
+                        cpulist_scnprintf(buf, len-2, mask) :
-                        cpumask_scnprintf(buf, len-2, *mask);
+                        cpumask_scnprintf(buf, len-2, mask);
                buf[n++] = '\n';
                buf[n] = '\0';
        }
diff --git a/drivers/clocksource/tcb_clksrc.c b/drivers/clocksource/tcb_clksrc.c
index f450588e5858..254f1064d973 100644
--- a/drivers/clocksource/tcb_clksrc.c
+++ b/drivers/clocksource/tcb_clksrc.c
@@ -154,7 +154,6 @@ static struct tc_clkevt_device clkevt = {
                .shift          = 32,
                /* Should be lower than at91rm9200's system timer */
                .rating         = 125,
-                .cpumask        = CPU_MASK_CPU0,
                .set_next_event = tc_next_event,
                .set_mode       = tc_mode,
        },
@@ -195,6 +194,7 @@ static void __init setup_clkevents(struct atmel_tc *tc, int clk32k_divisor_idx)
        clkevt.clkevt.max_delta_ns
                = clockevent_delta2ns(0xffff, &clkevt.clkevt);
        clkevt.clkevt.min_delta_ns = clockevent_delta2ns(1, &clkevt.clkevt) + 1;
+        clkevt.clkevt.cpumask = cpumask_of(0);
        setup_irq(irq, &tc_irqaction);
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++) {
+        for_each_online_cpu(i)
-                if (cpu_online(i))
+                destroy_comp_task(pool, 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) &&
+        if (!cpumask_empty(&current->cpus_allowed) &&
-            !cpus_full(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/drivers/parisc/iosapic.c b/drivers/parisc/iosapic.c
index 7beffcab2745..9dedbbd218c3 100644
--- a/drivers/parisc/iosapic.c
+++ b/drivers/parisc/iosapic.c
@@ -704,16 +704,17 @@ static unsigned int iosapic_startup_irq(unsigned int irq)
 }
 #ifdef CONFIG_SMP
-static void iosapic_set_affinity_irq(unsigned int irq, cpumask_t dest)
+static void iosapic_set_affinity_irq(unsigned int irq,
+                                     const struct cpumask *dest)
 {
        struct vector_info *vi = iosapic_get_vector(irq);
        u32 d0, d1, dummy_d0;
        unsigned long flags;
-        if (cpu_check_affinity(irq, &dest))
+        if (cpu_check_affinity(irq, dest))
                return;
-        vi->txn_addr = txn_affinity_addr(irq, first_cpu(dest));
+        vi->txn_addr = txn_affinity_addr(irq, cpumask_first(dest));
        spin_lock_irqsave(&iosapic_lock, flags);
        /* d1 contains the destination CPU, so only want to set that
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 5d72866897a8..c88485860a0a 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -74,7 +74,7 @@ static ssize_t local_cpus_show(struct device *dev,
        int len;
        mask = pcibus_to_cpumask(to_pci_dev(dev)->bus);
-        len = cpumask_scnprintf(buf, PAGE_SIZE-2, mask);
+        len = cpumask_scnprintf(buf, PAGE_SIZE-2, &mask);
        buf[len++] = '\n';
        buf[len] = '\0';
        return len;
@@ -88,7 +88,7 @@ static ssize_t local_cpulist_show(struct device *dev,
        int len;
        mask = pcibus_to_cpumask(to_pci_dev(dev)->bus);
-        len = cpulist_scnprintf(buf, PAGE_SIZE-2, mask);
+        len = cpulist_scnprintf(buf, PAGE_SIZE-2, &mask);
        buf[len++] = '\n';
        buf[len] = '\0';
        return len;
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 003a9b3c293f..5b3f5937ecf5 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -55,8 +55,8 @@ static ssize_t pci_bus_show_cpuaffinity(struct device *dev,
        cpumask = pcibus_to_cpumask(to_pci_bus(dev));
        ret = type?
-                cpulist_scnprintf(buf, PAGE_SIZE-2, cpumask):
+                cpulist_scnprintf(buf, PAGE_SIZE-2, &cpumask) :
-                cpumask_scnprintf(buf, PAGE_SIZE-2, cpumask);
+                cpumask_scnprintf(buf, PAGE_SIZE-2, &cpumask);
        buf[ret++] = '\n';
        buf[ret] = '\0';
        return ret;
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 46625cd38743..add640ff5c6c 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -588,7 +588,7 @@ void rebind_evtchn_irq(int evtchn, int irq)
        spin_unlock(&irq_mapping_update_lock);
        /* new event channels are always bound to cpu 0 */
-        irq_set_affinity(irq, cpumask_of_cpu(0));
+        irq_set_affinity(irq, cpumask_of(0));
        /* Unmask the event channel. */
        enable_irq(irq);
@@ -617,9 +617,9 @@ static void rebind_irq_to_cpu(unsigned irq, unsigned tcpu)
 }
-static void set_affinity_irq(unsigned irq, cpumask_t dest)
+static void set_affinity_irq(unsigned irq, const struct cpumask *dest)
 {
-        unsigned tcpu = first_cpu(dest);
+        unsigned tcpu = cpumask_first(dest);
        rebind_irq_to_cpu(irq, tcpu);
 }
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/asm-generic/topology.h b/include/asm-generic/topology.h
index 54bbf6e04ee8..0e9e2bc0ee96 100644
--- a/include/asm-generic/topology.h
+++ b/include/asm-generic/topology.h
@@ -40,6 +40,9 @@
 #ifndef node_to_cpumask
 #define node_to_cpumask(node)   ((void)node, cpu_online_map)
 #endif
+#ifndef cpumask_of_node
+#define cpumask_of_node(node)   ((void)node, cpu_online_mask)
+#endif
 #ifndef node_to_first_cpu
 #define node_to_first_cpu(node) ((void)(node),0)
 #endif
@@ -54,9 +57,18 @@
                                )
 #endif
+#ifndef cpumask_of_pcibus
+#define cpumask_of_pcibus(bus)  (pcibus_to_node(bus) == -1 ?            \
+                                 cpu_all_mask :                         \
+                                 cpumask_of_node(pcibus_to_node(bus)))
+#endif
 #endif  /* CONFIG_NUMA */
-/* returns pointer to cpumask for specified node */
+/*
+ * returns pointer to cpumask for specified node
+ * Deprecated: use "const struct cpumask *mask = cpumask_of_node(node)"
+ */
 #ifndef node_to_cpumask_ptr
 #define node_to_cpumask_ptr(v, node)                                    \
diff --git a/include/asm-m32r/smp.h b/include/asm-m32r/smp.h
index c5dd66916692..b96a6d2ffbc3 100644
--- a/include/asm-m32r/smp.h
+++ b/include/asm-m32r/smp.h
@@ -63,8 +63,6 @@ extern volatile int cpu_2_physid[NR_CPUS];
 #define raw_smp_processor_id()  (current_thread_info()->cpu)
 extern cpumask_t cpu_callout_map;
-extern cpumask_t cpu_possible_map;
-extern cpumask_t cpu_present_map;
 static __inline__ int hard_smp_processor_id(void)
 {
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/clockchips.h b/include/linux/clockchips.h
index ed3a5d473e52..cea153697ec7 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -82,13 +82,13 @@ struct clock_event_device {
        int                     shift;
        int                     rating;
        int                     irq;
-        cpumask_t               cpumask;
+        const struct cpumask    *cpumask;
        int                     (*set_next_event)(unsigned long evt,
                                                  struct clock_event_device *);
        void                    (*set_mode)(enum clock_event_mode mode,
                                            struct clock_event_device *);
        void                    (*event_handler)(struct clock_event_device *);
-        void                    (*broadcast)(cpumask_t mask);
+        void                    (*broadcast)(const struct cpumask *mask);
        struct list_head        list;
        enum clock_event_mode   mode;
        ktime_t                 next_event;
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 21e1dd43e52a..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);
 }
 /*
@@ -339,36 +358,6 @@ extern cpumask_t cpu_mask_all;
 #endif
 #define CPUMASK_PTR(v, m)       cpumask_t *v = &(m->v)
-#define cpumask_scnprintf(buf, len, src) \
-                        __cpumask_scnprintf((buf), (len), &(src), NR_CPUS)
-static inline int __cpumask_scnprintf(char *buf, int len,
-                                        const cpumask_t *srcp, int nbits)
-{
-        return bitmap_scnprintf(buf, len, srcp->bits, nbits);
-}
-#define cpumask_parse_user(ubuf, ulen, dst) \
-                        __cpumask_parse_user((ubuf), (ulen), &(dst), NR_CPUS)
-static inline int __cpumask_parse_user(const char __user *buf, int len,
-                                        cpumask_t *dstp, int nbits)
-{
-        return bitmap_parse_user(buf, len, dstp->bits, nbits);
-}
-#define cpulist_scnprintf(buf, len, src) \
-                        __cpulist_scnprintf((buf), (len), &(src), NR_CPUS)
-static inline int __cpulist_scnprintf(char *buf, int len,
-                                        const cpumask_t *srcp, int nbits)
-{
-        return bitmap_scnlistprintf(buf, len, srcp->bits, nbits);
-}
-#define cpulist_parse(buf, dst) __cpulist_parse((buf), &(dst), NR_CPUS)
-static inline int __cpulist_parse(const char *buf, cpumask_t *dstp, int nbits)
-{
-        return bitmap_parselist(buf, dstp->bits, nbits);
-}
 #define cpu_remap(oldbit, old, new) \
                __cpu_remap((oldbit), &(old), &(new), NR_CPUS)
 static inline int __cpu_remap(int oldbit,
@@ -446,74 +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
+ * possible, present, active and online cpus.
- * bitmap of size NR_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
+ *  If !CONFIG_HOTPLUG_CPU, present == possible, and active == online.
- *     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
 *
- *  In either case, NR_CPUS is fixed at compile time, as the static
+ *  The cpu_possible_mask is fixed at boot time, as the set of CPU id's
- *  size of these bitmaps.  The cpu_possible_map is fixed at boot
+ *  that it is possible might ever be plugged in at anytime during the
- *  time, as the set of CPU id's that it is possible might ever
+ *  life of that system boot.  The cpu_present_mask is dynamic(*),
- *  be plugged in at anytime during the life of that system boot.
+ *  representing which CPUs are currently plugged in.  And
- *  The cpu_present_map is dynamic(*), representing which CPUs
+ *  cpu_online_mask is the dynamic subset of cpu_present_mask,
- *  are currently plugged in.  And cpu_online_map is the dynamic
+ *  indicating those CPUs available for scheduling.
- *  subset of cpu_present_map, 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
+ *  (*) Well, cpu_present_mask is dynamic in the hotplug case.  If not
- *      hotplug, it's a copy of cpu_possible_map, hence fixed at boot.
+ *      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 const struct cpumask *const cpu_possible_mask;
-extern cpumask_t cpu_online_map;
+extern const struct cpumask *const cpu_online_mask;
-extern cpumask_t cpu_present_map;
+extern const struct cpumask *const cpu_present_mask;
-extern cpumask_t cpu_active_map;
+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_online_cpus()       cpumask_weight(cpu_online_mask)
-#define num_possible_cpus()     cpus_weight_nr(cpu_possible_map)
+#define num_possible_cpus()     cpumask_weight(cpu_possible_mask)
-#define num_present_cpus()      cpus_weight_nr(cpu_present_map)
+#define num_present_cpus()      cpumask_weight(cpu_present_mask)
-#define cpu_online(cpu)         cpu_isset((cpu), cpu_online_map)
+#define cpu_online(cpu)         cpumask_test_cpu((cpu), cpu_online_mask)
-#define cpu_possible(cpu)       cpu_isset((cpu), cpu_possible_map)
+#define cpu_possible(cpu)       cpumask_test_cpu((cpu), cpu_possible_mask)
-#define cpu_present(cpu)        cpu_isset((cpu), cpu_present_map)
+#define cpu_present(cpu)        cpumask_test_cpu((cpu), cpu_present_mask)
-#define cpu_active(cpu)         cpu_isset((cpu), cpu_active_map)
+#define cpu_active(cpu)         cpumask_test_cpu((cpu), cpu_active_mask)
 #else
 #define num_online_cpus()       1
 #define num_possible_cpus()     1
@@ -526,10 +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)
@@ -540,9 +514,6 @@ extern cpumask_t cpu_active_map;
        [BITS_TO_LONGS(NR_CPUS)-1] = CPU_MASK_LAST_WORD \
 }
-/* This produces more efficient code. */
-#define nr_cpumask_bits NR_CPUS
 #else /* NR_CPUS > BITS_PER_LONG */
 #define CPU_BITS_ALL                                            \
@@ -550,9 +521,15 @@ extern cpumask_t cpu_active_map;
        [0 ... BITS_TO_LONGS(NR_CPUS)-2] = ~0UL,                \
        [BITS_TO_LONGS(NR_CPUS)-1] = CPU_MASK_LAST_WORD         \
 }
+#endif /* NR_CPUS > BITS_PER_LONG */
+#ifdef CONFIG_CPUMASK_OFFSTACK
+/* Assuming NR_CPUS is huge, a runtime limit is more efficient.  Also,
+ * not all bits may be allocated. */
 #define nr_cpumask_bits nr_cpu_ids
-#endif /* NR_CPUS > BITS_PER_LONG */
+#else
+#define nr_cpumask_bits NR_CPUS
+#endif
 /* verify cpu argument to cpumask_* operators */
 static inline unsigned int cpumask_check(unsigned int cpu)
@@ -714,7 +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
@@ -946,22 +923,61 @@ static inline void cpumask_copy(struct cpumask *dstp,
 #define cpumask_of(cpu) (get_cpu_mask(cpu))
 /**
- * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
+ * cpumask_scnprintf - print a cpumask into a string as comma-separated hex
- * @bitmap: the bitmap
+ * @buf: the buffer to sprintf into
+ * @len: the length of the buffer
+ * @srcp: the cpumask to print
 *
- * There are a few places where cpumask_var_t isn't appropriate and
+ * If len is zero, returns zero.  Otherwise returns the length of the
- * static cpumasks must be used (eg. very early boot), yet we don't
+ * (nul-terminated) @buf string.
- * expose the definition of 'struct cpumask'.
+ */
+static inline int cpumask_scnprintf(char *buf, int len,
+                                    const struct cpumask *srcp)
+{
+        return bitmap_scnprintf(buf, len, cpumask_bits(srcp), nr_cpumask_bits);
+}
+/**
+ * cpumask_parse_user - extract a cpumask from a user string
+ * @buf: the buffer to extract from
+ * @len: the length of the buffer
+ * @dstp: the cpumask to set.
 *
- * This does the conversion, and can be used as a constant initializer.
+ * Returns -errno, or 0 for success.
 */
-#define to_cpumask(bitmap)                                              \
+static inline int cpumask_parse_user(const char __user *buf, int len,
-        ((struct cpumask *)(1 ? (bitmap)                                \
+                                     struct cpumask *dstp)
-                            : (void *)sizeof(__check_is_bitmap(bitmap))))
+{
+        return bitmap_parse_user(buf, len, cpumask_bits(dstp), nr_cpumask_bits);
+}
-static inline int __check_is_bitmap(const unsigned long *bitmap)
+/**
+ * cpulist_scnprintf - print a cpumask into a string as comma-separated list
+ * @buf: the buffer to sprintf into
+ * @len: the length of the buffer
+ * @srcp: the cpumask to print
+ *
+ * If len is zero, returns zero.  Otherwise returns the length of the
+ * (nul-terminated) @buf string.
+ */
+static inline int cpulist_scnprintf(char *buf, int len,
+                                    const struct cpumask *srcp)
 {
-        return 1;
+        return bitmap_scnlistprintf(buf, len, cpumask_bits(srcp),
+                                    nr_cpumask_bits);
+}
+/**
+ * cpulist_parse_user - extract a cpumask from a user string of ranges
+ * @buf: the buffer to extract from
+ * @len: the length of the buffer
+ * @dstp: the cpumask to set.
+ *
+ * Returns -errno, or 0 for success.
+ */
+static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
+{
+        return bitmap_parselist(buf, cpumask_bits(dstp), nr_cpumask_bits);
 }
 /**
@@ -995,6 +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);
@@ -1008,6 +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)
 {
 }
@@ -1021,12 +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);
@@ -1035,51 +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 */
+#define for_each_possible_cpu(cpu) for_each_cpu((cpu), cpu_possible_mask)
-static inline void set_cpu_possible(unsigned int cpu, bool possible)
+#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)
-        if (possible)
-                cpumask_set_cpu(cpu, &cpu_possible_map);
-        else
-                cpumask_clear_cpu(cpu, &cpu_possible_map);
-}
-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)
+/* Wrappers for arch boot code to manipulate normally-constant masks */
-{
+void set_cpu_possible(unsigned int cpu, bool possible);
-        cpumask_copy(&cpu_online_map, src);
+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/interrupt.h b/include/linux/interrupt.h
index be3c484b5242..dfaee6bd265b 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -111,13 +111,13 @@ extern void enable_irq(unsigned int irq);
 extern cpumask_t irq_default_affinity;
-extern int irq_set_affinity(unsigned int irq, cpumask_t cpumask);
+extern int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask);
 extern int irq_can_set_affinity(unsigned int irq);
 extern int irq_select_affinity(unsigned int irq);
 #else /* CONFIG_SMP */
-static inline int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
+static inline int irq_set_affinity(unsigned int irq, const struct cpumask *m)
 {
        return -EINVAL;
 }
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 98564dc64476..5845bdc1ac09 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -113,7 +113,8 @@ struct irq_chip {
        void            (*eoi)(unsigned int irq);
        void            (*end)(unsigned int irq);
-        void            (*set_affinity)(unsigned int irq, cpumask_t dest);
+        void            (*set_affinity)(unsigned int irq,
+                                        const struct cpumask *dest);
        int             (*retrigger)(unsigned int irq);
        int             (*set_type)(unsigned int irq, unsigned int flow_type);
        int             (*set_wake)(unsigned int irq, unsigned int on);
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);
+int seq_bitmap(struct seq_file *m, const unsigned long *bits,
-static inline int seq_cpumask(struct seq_file *m, cpumask_t *mask)
+                                   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. */
+void smp_call_function_many(const struct cpumask *mask,
-int smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
+                            void (*func)(void *info), void *info, bool wait);
-                                int wait);
-static inline void smp_call_function_many(const struct cpumask *mask,
+/* Deprecated: Use smp_call_function_many which takes a pointer to the mask. */
-                                          void (*func)(void *info), void *info,
+static inline int
-                                          int wait)
+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
+ * Maximum supported processors.  Setting this smaller saves quite a
- * set via configure setting.  The maximum is equal to the size of the
+ * bit of memory.  Use nr_cpu_ids instead of this except for static bitmaps.
- * bitmasks used on that platform, i.e. 32 or 64.  Setting this smaller
- * saves quite a bit of memory.
 */
-#ifdef CONFIG_SMP
+#ifndef CONFIG_NR_CPUS
-#define NR_CPUS         CONFIG_NR_CPUS
+/* FIXME: This should be fixed in the arch's Kconfig */
-#else
+#define CONFIG_NR_CPUS  1
-#define 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/init/Kconfig b/init/Kconfig
index 13627191a60d..f6281711166d 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -924,6 +924,15 @@ config KMOD
 endif # MODULES
+config INIT_ALL_POSSIBLE
+        bool
+        help
+          Back when each arch used to define their own cpu_online_map and
+          cpu_possible_map, some of them chose to initialize cpu_possible_map
+          with all 1s, and others with all 0s.  When they were centralised,
+          it was better to provide this option than to break all the archs
+          and have several arch maintainers persuing me down dark alleys.
 config STOP_MACHINE
        bool
        default y
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 8ea32e8d68b0..2c9f78f3a2fc 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -15,29 +15,8 @@
 #include <linux/stop_machine.h>
 #include <linux/mutex.h>
-/*
+#ifdef CONFIG_SMP
- * Represents all cpu's present in the system
+/* Serializes the updates to cpu_online_mask, cpu_present_mask */
- * 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);
-#ifndef CONFIG_SMP
-/*
- * Represents all cpu's that are currently online.
- */
-cpumask_t cpu_online_map __read_mostly = CPU_MASK_ALL;
-EXPORT_SYMBOL(cpu_online_map);
-cpumask_t cpu_possible_map __read_mostly = CPU_MASK_ALL;
-EXPORT_SYMBOL(cpu_possible_map);
-#else /* CONFIG_SMP */
-/* Serializes the updates to cpu_online_map, cpu_present_map */
 static DEFINE_MUTEX(cpu_add_remove_lock);
 static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain);
@@ -64,8 +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)
@@ -96,7 +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)
 {
@@ -502,3 +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/cpuset.c b/kernel/cpuset.c
index 96c0ba13b8cd..39c1a4c1c5a9 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -896,7 +896,7 @@ static int update_cpumask(struct cpuset *cs, const char *buf)
        if (!*buf) {
                cpus_clear(trialcs.cpus_allowed);
        } else {
-                retval = cpulist_parse(buf, trialcs.cpus_allowed);
+                retval = cpulist_parse(buf, &trialcs.cpus_allowed);
                if (retval < 0)
                        return retval;
@@ -1482,7 +1482,7 @@ static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs)
        mask = cs->cpus_allowed;
        mutex_unlock(&callback_mutex);
-        return cpulist_scnprintf(page, PAGE_SIZE, mask);
+        return cpulist_scnprintf(page, PAGE_SIZE, &mask);
 }
 static int cpuset_sprintf_memlist(char *page, struct cpuset *cs)
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 6eb3c7952b64..f63c706d25e1 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -46,7 +46,7 @@ void dynamic_irq_init(unsigned int irq)
        desc->irq_count = 0;
        desc->irqs_unhandled = 0;
 #ifdef CONFIG_SMP
-        cpus_setall(desc->affinity);
+        cpumask_setall(&desc->affinity);
 #endif
        spin_unlock_irqrestore(&desc->lock, flags);
 }
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 540f6c49f3fa..61c4a9b62165 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -79,7 +79,7 @@ int irq_can_set_affinity(unsigned int irq)
 *      @cpumask:       cpumask
 *
 */
-int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
+int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
        struct irq_desc *desc = irq_to_desc(irq);
        unsigned long flags;
@@ -91,14 +91,14 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
 #ifdef CONFIG_GENERIC_PENDING_IRQ
        if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) {
-                desc->affinity = cpumask;
+                cpumask_copy(&desc->affinity, cpumask);
                desc->chip->set_affinity(irq, cpumask);
        } else {
                desc->status |= IRQ_MOVE_PENDING;
-                desc->pending_mask = cpumask;
+                cpumask_copy(&desc->pending_mask, cpumask);
        }
 #else
-        desc->affinity = cpumask;
+        cpumask_copy(&desc->affinity, cpumask);
        desc->chip->set_affinity(irq, cpumask);
 #endif
        desc->status |= IRQ_AFFINITY_SET;
@@ -112,26 +112,24 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
 */
 int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc)
 {
-        cpumask_t mask;
        if (!irq_can_set_affinity(irq))
                return 0;
-        cpus_and(mask, cpu_online_map, irq_default_affinity);
        /*
         * Preserve an userspace affinity setup, but make sure that
         * one of the targets is online.
         */
        if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) {
-                if (cpus_intersects(desc->affinity, cpu_online_map))
+                if (cpumask_any_and(&desc->affinity, cpu_online_mask)
-                        mask = desc->affinity;
+                    < nr_cpu_ids)
+                        goto set_affinity;
                else
                        desc->status &= ~IRQ_AFFINITY_SET;
        }
-        desc->affinity = mask;
+        cpumask_and(&desc->affinity, cpu_online_mask, &irq_default_affinity);
-        desc->chip->set_affinity(irq, mask);
+set_affinity:
+        desc->chip->set_affinity(irq, &desc->affinity);
        return 0;
 }
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 9db681d95814..bd72329e630c 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -4,7 +4,6 @@
 void move_masked_irq(int irq)
 {
        struct irq_desc *desc = irq_to_desc(irq);
-        cpumask_t tmp;
        if (likely(!(desc->status & IRQ_MOVE_PENDING)))
                return;
@@ -19,7 +18,7 @@ void move_masked_irq(int irq)
        desc->status &= ~IRQ_MOVE_PENDING;
-        if (unlikely(cpus_empty(desc->pending_mask)))
+        if (unlikely(cpumask_empty(&desc->pending_mask)))
                return;
        if (!desc->chip->set_affinity)
@@ -27,8 +26,6 @@ void move_masked_irq(int irq)
        assert_spin_locked(&desc->lock);
-        cpus_and(tmp, desc->pending_mask, cpu_online_map);
        /*
         * If there was a valid mask to work with, please
         * do the disable, re-program, enable sequence.
@@ -41,10 +38,13 @@ void move_masked_irq(int irq)
         * For correct operation this depends on the caller
         * masking the irqs.
         */
-        if (likely(!cpus_empty(tmp))) {
+        if (likely(cpumask_any_and(&desc->pending_mask, cpu_online_mask)
-                desc->chip->set_affinity(irq,tmp);
+                   < nr_cpu_ids)) {
+                cpumask_and(&desc->affinity,
+                            &desc->pending_mask, cpu_online_mask);
+                desc->chip->set_affinity(irq, &desc->affinity);
        }
-        cpus_clear(desc->pending_mask);
+        cpumask_clear(&desc->pending_mask);
 }
 void move_native_irq(int irq)
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index f6b3440f05bc..d2c0e5ee53c5 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -40,33 +40,42 @@ static ssize_t irq_affinity_proc_write(struct file *file,
                const char __user *buffer, size_t count, loff_t *pos)
 {
        unsigned int irq = (int)(long)PDE(file->f_path.dentry->d_inode)->data;
-        cpumask_t new_value;
+        cpumask_var_t new_value;
        int err;
        if (!irq_to_desc(irq)->chip->set_affinity || no_irq_affinity ||
            irq_balancing_disabled(irq))
                return -EIO;
+        if (!alloc_cpumask_var(&new_value, GFP_KERNEL))
+                return -ENOMEM;
        err = cpumask_parse_user(buffer, count, new_value);
        if (err)
-                return err;
+                goto free_cpumask;
-        if (!is_affinity_mask_valid(new_value))
+        if (!is_affinity_mask_valid(*new_value)) {
-                return -EINVAL;
+                err = -EINVAL;
+                goto free_cpumask;
+        }
        /*
         * Do not allow disabling IRQs completely - it's a too easy
         * way to make the system unusable accidentally :-) At least
         * one online CPU still has to be targeted.
         */
-        if (!cpus_intersects(new_value, cpu_online_map))
+        if (!cpumask_intersects(new_value, cpu_online_mask)) {
                /* Special case for empty set - allow the architecture
                   code to set default SMP affinity. */
-                return irq_select_affinity_usr(irq) ? -EINVAL : count;
+                err = irq_select_affinity_usr(irq) ? -EINVAL : count;
+        } else {
-        irq_set_affinity(irq, new_value);
+                irq_set_affinity(irq, new_value);
+                err = count;
+        }
-        return count;
+free_cpumask:
+        free_cpumask_var(new_value);
+        return err;
 }
 static int irq_affinity_proc_open(struct inode *inode, struct file *file)
@@ -95,7 +104,7 @@ static ssize_t default_affinity_write(struct file *file,
        cpumask_t new_value;
        int err;
-        err = cpumask_parse_user(buffer, count, new_value);
+        err = cpumask_parse_user(buffer, count, &new_value);
        if (err)
                return err;
diff --git a/kernel/profile.c b/kernel/profile.c
index 60adefb59b5e..4cb7d68fed82 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -442,7 +442,7 @@ void profile_tick(int type)
 static int prof_cpu_mask_read_proc(char *page, char **start, off_t off,
                        int count, int *eof, void *data)
 {
-        int len = cpumask_scnprintf(page, count, *(cpumask_t *)data);
+        int len = cpumask_scnprintf(page, count, (cpumask_t *)data);
        if (count - len < 2)
                return -EINVAL;
        len += sprintf(page + len, "\n");
@@ -456,7 +456,7 @@ static int prof_cpu_mask_write_proc(struct file *file,
        unsigned long full_count = count, err;
        cpumask_t new_value;
-        err = cpumask_parse_user(buffer, count, new_value);
+        err = cpumask_parse_user(buffer, count, &new_value);
        if (err)
                return err;
diff --git a/kernel/sched.c b/kernel/sched.c
index fff1c4a20b65..f2095660efec 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6644,7 +6644,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
        struct sched_group *group = sd->groups;
        char str[256];
-        cpulist_scnprintf(str, sizeof(str), sd->span);
+        cpulist_scnprintf(str, sizeof(str), &sd->span);
        cpus_clear(*groupmask);
        printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
@@ -6697,7 +6697,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
                cpus_or(*groupmask, *groupmask, group->cpumask);
-                cpulist_scnprintf(str, sizeof(str), group->cpumask);
+                cpulist_scnprintf(str, sizeof(str), &group->cpumask);
                printk(KERN_CONT " %s", str);
                group = group->next;
@@ -7098,7 +7098,7 @@ cpu_to_phys_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
 {
        int group;
 #ifdef CONFIG_SCHED_MC
-        *mask = cpu_coregroup_map(cpu);
+        *mask = *cpu_coregroup_mask(cpu);
        cpus_and(*mask, *mask, *cpu_map);
        group = first_cpu(*mask);
 #elif defined(CONFIG_SCHED_SMT)
@@ -7471,7 +7471,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
                sd = &per_cpu(core_domains, i);
                SD_INIT(sd, MC);
                set_domain_attribute(sd, attr);
-                sd->span = cpu_coregroup_map(i);
+                sd->span = *cpu_coregroup_mask(i);
                cpus_and(sd->span, sd->span, *cpu_map);
                sd->parent = p;
                p->child = sd;
@@ -7514,7 +7514,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
                SCHED_CPUMASK_VAR(this_core_map, allmasks);
                SCHED_CPUMASK_VAR(send_covered, allmasks);
-                *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 != first_cpu(*this_core_map))
                        continue;
diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h
index 3b01098164c8..b59fd9cdc1b6 100644
--- a/kernel/sched_stats.h
+++ b/kernel/sched_stats.h
@@ -42,7 +42,7 @@ static int show_schedstat(struct seq_file *seq, void *v)
                for_each_domain(cpu, sd) {
                        enum cpu_idle_type itype;
-                        cpumask_scnprintf(mask_str, mask_len, sd->span);
+                        cpumask_scnprintf(mask_str, mask_len, &sd->span);
                        seq_printf(seq, "domain%d %s", dcount++, mask_str);
                        for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES;
                                        itype++) {
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 */
+/* FIXME: Shim for archs using old arch_send_call_function_ipi API. */
-static void quiesce_dummy(void *unused)
+#ifndef arch_send_call_function_ipi_mask
-{
+#define arch_send_call_function_ipi_mask(maskp) \
-}
+        arch_send_call_function_ipi(*(maskp))
+#endif
-/*
- * 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);
-        }
-}
 /**
- * smp_call_function_mask(): Run a function on a set of other CPUs.
+ * smp_call_function_many(): Run a function on a set of other CPUs.
- * @mask: The set of cpus to run on.
+ * @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,
+void smp_call_function_many(const struct cpumask *mask,
-                           int wait)
+                            void (*func)(void *), void *info,
+                            bool wait)
 {
-        struct call_function_data d;
+        struct call_function_data *data;
-        struct call_function_data *data = NULL;
-        cpumask_t allbutself;
        unsigned long flags;
-        int cpu, num_cpus;
+        int cpu, next_cpu;
-        int slowpath = 0;
        /* Can deadlock when called with interrupts disabled */
        WARN_ON(irqs_disabled());
-        cpu = smp_processor_id();
+        /* So, what's a CPU they want?  Ignoring this one. */
-        allbutself = cpu_online_map;
+        cpu = cpumask_first_and(mask, cpu_online_mask);
-        cpu_clear(cpu, allbutself);
+        if (cpu == smp_processor_id())
-        cpus_and(mask, mask, allbutself);
+                cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
-        num_cpus = cpus_weight(mask);
+        /* No online cpus?  We're done. */
+        if (cpu >= nr_cpu_ids)
-        /*
+                return;
-         * If zero CPUs, return. If just a single CPU, turn this request
-         * into a targetted single call instead since it's faster.
+        /* Do we have another CPU which isn't us? */
-         */
+        next_cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
-        if (!num_cpus)
+        if (next_cpu == smp_processor_id())
-                return 0;
+                next_cpu = cpumask_next_and(next_cpu, mask, cpu_online_mask);
-        else if (num_cpus == 1) {
-                cpu = first_cpu(mask);
+        /* Fastpath: do that cpu by itself. */
-                return smp_call_function_single(cpu, func, info, wait);
+        if (next_cpu >= nr_cpu_ids) {
+                smp_call_function_single(cpu, func, info, wait);
+                return;
        }
-        data = kmalloc(sizeof(*data), GFP_ATOMIC);
+        data = kmalloc(sizeof(*data) + cpumask_size(), GFP_ATOMIC);
-        if (data) {
+        if (unlikely(!data)) {
-                data->csd.flags = CSD_FLAG_ALLOC;
+                /* Slow path. */
-                if (wait)
+                for_each_online_cpu(cpu) {
-                        data->csd.flags |= CSD_FLAG_WAIT;
+                        if (cpu == smp_processor_id())
-        } else {
+                                continue;
-                data = &d;
+                        if (cpumask_test_cpu(cpu, mask))
-                data->csd.flags = CSD_FLAG_WAIT;
+                                smp_call_function_single(cpu, func, info, wait);
-                wait = 1;
+                }
-                slowpath = 1;
+                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;
+        cpumask_and(to_cpumask(data->cpumask_bits), mask, cpu_online_mask);
-        data->cpumask = 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/kernel/taskstats.c b/kernel/taskstats.c
index bd6be76303cf..6d7dc4ec4aa5 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -352,7 +352,7 @@ static int parse(struct nlattr *na, cpumask_t *mask)
        if (!data)
                return -ENOMEM;
        nla_strlcpy(data, na, len);
-        ret = cpulist_parse(data, *mask);
+        ret = cpulist_parse(data, mask);
        kfree(data);
        return ret;
 }
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index f8d968063cea..ea2f48af83cf 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -166,6 +166,8 @@ static void clockevents_notify_released(void)
 void clockevents_register_device(struct clock_event_device *dev)
 {
        BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
+        BUG_ON(!dev->cpumask);
        /*
         * A nsec2cyc multiplicator of 0 is invalid and we'd crash
         * on it, so fix it up and emit a warning:
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index f98a1b7b16e9..9590af2327be 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -150,7 +150,7 @@ static void tick_do_broadcast(cpumask_t mask)
                 */
                cpu = first_cpu(mask);
                td = &per_cpu(tick_cpu_device, cpu);
-                td->evtdev->broadcast(mask);
+                td->evtdev->broadcast(&mask);
        }
 }
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index df12434b43ca..f8372be74122 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -136,7 +136,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
 */
 static void tick_setup_device(struct tick_device *td,
                              struct clock_event_device *newdev, int cpu,
-                              const cpumask_t *cpumask)
+                              const struct cpumask *cpumask)
 {
        ktime_t next_event;
        void (*handler)(struct clock_event_device *) = NULL;
@@ -171,8 +171,8 @@ static void tick_setup_device(struct tick_device *td,
         * When the device is not per cpu, pin the interrupt to the
         * current cpu:
         */
-        if (!cpus_equal(newdev->cpumask, *cpumask))
+        if (!cpumask_equal(newdev->cpumask, cpumask))
-                irq_set_affinity(newdev->irq, *cpumask);
+                irq_set_affinity(newdev->irq, cpumask);
        /*
         * When global broadcasting is active, check if the current
@@ -202,14 +202,14 @@ static int tick_check_new_device(struct clock_event_device *newdev)
        spin_lock_irqsave(&tick_device_lock, flags);
        cpu = smp_processor_id();
-        if (!cpu_isset(cpu, newdev->cpumask))
+        if (!cpumask_test_cpu(cpu, newdev->cpumask))
                goto out_bc;
        td = &per_cpu(tick_cpu_device, cpu);
        curdev = td->evtdev;
        /* cpu local device ? */
-        if (!cpus_equal(newdev->cpumask, cpumask_of_cpu(cpu))) {
+        if (!cpumask_equal(newdev->cpumask, cpumask_of(cpu))) {
                /*
                 * If the cpu affinity of the device interrupt can not
@@ -222,7 +222,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
                 * If we have a cpu local device already, do not replace it
                 * by a non cpu local device
                 */
-                if (curdev && cpus_equal(curdev->cpumask, cpumask_of_cpu(cpu)))
+                if (curdev && cpumask_equal(curdev->cpumask, cpumask_of(cpu)))
                        goto out_bc;
        }
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 4185d5221633..0e91f43b6baf 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -2674,7 +2674,7 @@ tracing_cpumask_read(struct file *filp, char __user *ubuf,
        mutex_lock(&tracing_cpumask_update_lock);
-        len = cpumask_scnprintf(mask_str, count, tracing_cpumask);
+        len = cpumask_scnprintf(mask_str, count, &tracing_cpumask);
        if (count - len < 2) {
                count = -EINVAL;
                goto out_err;
@@ -2695,7 +2695,7 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf,
        int err, cpu;
        mutex_lock(&tracing_cpumask_update_lock);
-        err = cpumask_parse_user(ubuf, count, tracing_cpumask_new);
+        err = cpumask_parse_user(ubuf, count, &tracing_cpumask_new);
        if (err)
                goto err_unlock;
diff --git a/lib/Kconfig b/lib/Kconfig
index fd4118e097f0..2ba43c4a5b07 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -159,4 +159,11 @@ config CHECK_SIGNATURE
 config HAVE_LMB
        boolean
+config CPUMASK_OFFSTACK
+        bool "Force CPU masks off stack" if DEBUG_PER_CPU_MAPS
+        help
+          Use dynamic allocation for cpumask_var_t, instead of putting
+          them on the stack.  This is a bit more expensive, but avoids
+          stack overflow.
 endmenu
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());
diff --git a/mm/slub.c b/mm/slub.c
index 6cb7ad107852..0d861c3154b6 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -3642,7 +3642,7 @@ static int list_locations(struct kmem_cache *s, char *buf,
                                len < PAGE_SIZE - 60) {
                        len += sprintf(buf + len, " cpus=");
                        len += cpulist_scnprintf(buf + len, PAGE_SIZE - len - 50,
-                                        l->cpus);
+                                        &l->cpus);
                }
                if (num_online_nodes() > 1 && !nodes_empty(l->nodes) &&