diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 82469db25100..5a19e2eb5711 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3514,7 +3514,7 @@ int __init blk_dev_init(void)
 	iocontext_cachep = kmem_cache_create("blkdev_ioc",
 			sizeof(struct io_context), 0, SLAB_PANIC, NULL, NULL);
 
-	for_each_cpu(i)
+	for_each_possible_cpu(i)
 		INIT_LIST_HEAD(&per_cpu(blk_cpu_done, i));
 
 	open_softirq(BLOCK_SOFTIRQ, blk_done_softirq, NULL);
diff --git a/fs/file.c b/fs/file.c
index bbc743314730..55f4e7022563 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -373,6 +373,6 @@ static void __devinit fdtable_defer_list_init(int cpu)
 void __init files_defer_init(void)
 {
 	int i;
-	for_each_cpu(i)
+	for_each_possible_cpu(i)
 		fdtable_defer_list_init(i);
 }
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 1e9ea37d457e..1edce0c34bfd 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -534,7 +534,7 @@ static int show_stat(struct seq_file *p, void *v)
 	if (wall_to_monotonic.tv_nsec)
 		--jif;
 
-	for_each_cpu(i) {
+	for_each_possible_cpu(i) {
 		int j;
 
 		user = cputime64_add(user, kstat_cpu(i).cpustat.user);
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index 78cf45547e31..c0caf433a7d7 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -19,7 +19,7 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
 	unsigned int __i;					\
-	for_each_cpu(__i)					\
+	for_each_possible_cpu(__i)				\
 		memcpy((pcpudst)+__per_cpu_offset[__i],		\
 		       (src), (size));				\
 } while (0)
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 3c1b0294a742..10a27f29d692 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -152,14 +152,14 @@ struct disk_attribute {
 ({									\
 	typeof(gendiskp->dkstats->field) res = 0;			\
 	int i;								\
-	for_each_cpu(i)							\
+	for_each_possible_cpu(i)					\
 		res += per_cpu_ptr(gendiskp->dkstats, i)->field;	\
 	res;								\
 })
 
 static inline void disk_stat_set_all(struct gendisk *gendiskp, int value)	{
 	int i;
-	for_each_cpu(i)
+	for_each_possible_cpu(i)
 		memset(per_cpu_ptr(gendiskp->dkstats, i), value,
 				sizeof (struct disk_stats));
 }		
diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index a484572c302e..b46249082cca 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -46,7 +46,7 @@ static inline int kstat_irqs(int irq)
 {
 	int cpu, sum = 0;
 
-	for_each_cpu(cpu)
+	for_each_possible_cpu(cpu)
 		sum += kstat_cpu(cpu).irqs[irq];
 
 	return sum;
diff --git a/init/main.c b/init/main.c
index 64466ea1984c..4a2f0898dda1 100644
--- a/init/main.c
+++ b/init/main.c
@@ -341,7 +341,7 @@ static void __init setup_per_cpu_areas(void)
 #endif
 	ptr = alloc_bootmem(size * nr_possible_cpus);
 
-	for_each_cpu(i) {
+	for_each_possible_cpu(i) {
 		__per_cpu_offset[i] = ptr - __per_cpu_start;
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 		ptr += size;
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index b4b362b5baf5..8154e7589d12 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -301,7 +301,7 @@ rcu_torture_printk(char *page)
 	long pipesummary[RCU_TORTURE_PIPE_LEN + 1] = { 0 };
 	long batchsummary[RCU_TORTURE_PIPE_LEN + 1] = { 0 };
 
-	for_each_cpu(cpu) {
+	for_each_possible_cpu(cpu) {
 		for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) {
 			pipesummary[i] += per_cpu(rcu_torture_count, cpu)[i];
 			batchsummary[i] += per_cpu(rcu_torture_batch, cpu)[i];
@@ -535,7 +535,7 @@ rcu_torture_init(void)
 	atomic_set(&n_rcu_torture_error, 0);
 	for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++)
 		atomic_set(&rcu_torture_wcount[i], 0);
-	for_each_cpu(cpu) {
+	for_each_possible_cpu(cpu) {
 		for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) {
 			per_cpu(rcu_torture_count, cpu)[i] = 0;
 			per_cpu(rcu_torture_batch, cpu)[i] = 0;
diff --git a/kernel/sched.c b/kernel/sched.c
index 7854ee516b92..a9ecac398bb9 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1625,7 +1625,7 @@ unsigned long nr_uninterruptible(void)
 {
 	unsigned long i, sum = 0;
 
-	for_each_cpu(i)
+	for_each_possible_cpu(i)
 		sum += cpu_rq(i)->nr_uninterruptible;
 
 	/*
@@ -1642,7 +1642,7 @@ unsigned long long nr_context_switches(void)
 {
 	unsigned long long i, sum = 0;
 
-	for_each_cpu(i)
+	for_each_possible_cpu(i)
 		sum += cpu_rq(i)->nr_switches;
 
 	return sum;
@@ -1652,7 +1652,7 @@ unsigned long nr_iowait(void)
 {
 	unsigned long i, sum = 0;
 
-	for_each_cpu(i)
+	for_each_possible_cpu(i)
 		sum += atomic_read(&cpu_rq(i)->nr_iowait);
 
 	return sum;
@@ -6080,7 +6080,7 @@ void __init sched_init(void)
 	runqueue_t *rq;
 	int i, j, k;
 
-	for_each_cpu(i) {
+	for_each_possible_cpu(i) {
 		prio_array_t *array;
 
 		rq = cpu_rq(i);
diff --git a/mm/slab.c b/mm/slab.c
index 681837499d7d..4cbf8bb13557 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3311,7 +3311,7 @@ void *__alloc_percpu(size_t size)
 	 * and we have no way of figuring out how to fix the array
 	 * that we have allocated then....
 	 */
-	for_each_cpu(i) {
+	for_each_possible_cpu(i) {
 		int node = cpu_to_node(i);
 
 		if (node_online(node))
@@ -3398,7 +3398,7 @@ void free_percpu(const void *objp)
 	/*
 	 * We allocate for all cpus so we cannot use for online cpu here.
 	 */
-	for_each_cpu(i)
+	for_each_possible_cpu(i)
 	    kfree(p->ptrs[i]);
 	kfree(p);
 }
diff --git a/mm/swap.c b/mm/swap.c
index 91b7e2026f69..88895c249bc9 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -512,7 +512,7 @@ long percpu_counter_sum(struct percpu_counter *fbc)
 
 	spin_lock(&fbc->lock);
 	ret = fbc->count;
-	for_each_cpu(cpu) {
+	for_each_possible_cpu(cpu) {
 		long *pcount = per_cpu_ptr(fbc->counters, cpu);
 		ret += *pcount;
 	}