8 files changed, 135 insertions, 26 deletions

diff --git a/mm/Makefile b/mm/Makefile
index f10c753dce6d..0b8f73f2ed16 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -10,7 +10,7 @@ mmu-$(CONFIG_MMU)	:= fremap.o highmem.o madvise.o memory.o mincore.o \
 obj-y                   := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
                            page_alloc.o page-writeback.o pdflush.o \
                            readahead.o swap.o truncate.o vmscan.o \
-                           prio_tree.o util.o $(mmu-y)
+                           prio_tree.o util.o mmzone.o $(mmu-y)
 
 obj-$(CONFIG_SWAP)      += page_io.o swap_state.o swapfile.o thrash.o
 obj-$(CONFIG_HUGETLBFS) += hugetlb.o
diff --git a/mm/bootmem.c b/mm/bootmem.c
index b55bd39fc5dd..d3e3bd2ffcea 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -33,6 +33,7 @@ EXPORT_SYMBOL(max_pfn);		/* This is exported so
                                  * dma_get_required_mask(), which uses
                                  * it, can be an inline function */
 
+static LIST_HEAD(bdata_list);
 #ifdef CONFIG_CRASH_DUMP
 /*
  * If we have booted due to a crash, max_pfn will be a very low value. We need
@@ -52,6 +53,27 @@ unsigned long __init bootmem_bootmap_pages (unsigned long pages)
 
         return mapsize;
 }
+/*
+ * link bdata in order
+ */
+static void link_bootmem(bootmem_data_t *bdata)
+{
+        bootmem_data_t *ent;
+        if (list_empty(&bdata_list)) {
+                list_add(&bdata->list, &bdata_list);
+                return;
+        }
+        /* insert in order */
+        list_for_each_entry(ent, &bdata_list, list) {
+                if (bdata->node_boot_start < ent->node_boot_start) {
+                        list_add_tail(&bdata->list, &ent->list);
+                        return;
+                }
+        }
+        list_add_tail(&bdata->list, &bdata_list);
+        return;
+}
+
 
 /*
  * Called once to set up the allocator itself.
@@ -62,13 +84,11 @@ static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
         bootmem_data_t *bdata = pgdat->bdata;
         unsigned long mapsize = ((end - start)+7)/8;
 
-        pgdat->pgdat_next = pgdat_list;
-        pgdat_list = pgdat;
-
         mapsize = ALIGN(mapsize, sizeof(long));
         bdata->node_bootmem_map = phys_to_virt(mapstart << PAGE_SHIFT);
         bdata->node_boot_start = (start << PAGE_SHIFT);
         bdata->node_low_pfn = end;
+        link_bootmem(bdata);
 
         /*
          * Initially all pages are reserved - setup_arch() has to
@@ -383,12 +403,11 @@ unsigned long __init free_all_bootmem (void)
 
 void * __init __alloc_bootmem(unsigned long size, unsigned long align, unsigned long goal)
 {
-        pg_data_t *pgdat = pgdat_list;
+        bootmem_data_t *bdata;
         void *ptr;
 
-        for_each_pgdat(pgdat)
-                if ((ptr = __alloc_bootmem_core(pgdat->bdata, size,
-                                                 align, goal, 0)))
+        list_for_each_entry(bdata, &bdata_list, list)
+                if ((ptr = __alloc_bootmem_core(bdata, size, align, goal, 0)))
                         return(ptr);
 
         /*
@@ -416,11 +435,11 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigne
 
 void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, unsigned long goal)
 {
-        pg_data_t *pgdat = pgdat_list;
+        bootmem_data_t *bdata;
         void *ptr;
 
-        for_each_pgdat(pgdat)
-                if ((ptr = __alloc_bootmem_core(pgdat->bdata, size,
+        list_for_each_entry(bdata, &bdata_list, list)
+                if ((ptr = __alloc_bootmem_core(bdata, size,
                                                  align, goal, LOW32LIMIT)))
                         return(ptr);
 
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 4f71cfd29c6f..dec8249e972d 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -912,7 +912,7 @@ asmlinkage long sys_migrate_pages(pid_t pid, unsigned long maxnode,
         /*
          * Check if this process has the right to modify the specified
          * process. The right exists if the process has administrative
-         * capabilities, superuser priviledges or the same
+         * capabilities, superuser privileges or the same
          * userid as the target process.
          */
         if ((current->euid != task->suid) && (current->euid != task->uid) &&
diff --git a/mm/mmzone.c b/mm/mmzone.c
new file mode 100644
index 000000000000..b022370e612e
--- /dev/null
+++ b/mm/mmzone.c
@@ -0,0 +1,50 @@
+/*
+ * linux/mm/mmzone.c
+ *
+ * management codes for pgdats and zones.
+ */
+
+
+#include <linux/config.h>
+#include <linux/stddef.h>
+#include <linux/mmzone.h>
+#include <linux/module.h>
+
+struct pglist_data *first_online_pgdat(void)
+{
+        return NODE_DATA(first_online_node);
+}
+
+EXPORT_SYMBOL(first_online_pgdat);
+
+struct pglist_data *next_online_pgdat(struct pglist_data *pgdat)
+{
+        int nid = next_online_node(pgdat->node_id);
+
+        if (nid == MAX_NUMNODES)
+                return NULL;
+        return NODE_DATA(nid);
+}
+EXPORT_SYMBOL(next_online_pgdat);
+
+
+/*
+ * next_zone - helper magic for for_each_zone()
+ */
+struct zone *next_zone(struct zone *zone)
+{
+        pg_data_t *pgdat = zone->zone_pgdat;
+
+        if (zone < pgdat->node_zones + MAX_NR_ZONES - 1)
+                zone++;
+        else {
+                pgdat = next_online_pgdat(pgdat);
+                if (pgdat)
+                        zone = pgdat->node_zones;
+                else
+                        zone = NULL;
+        }
+        return zone;
+}
+EXPORT_SYMBOL(next_zone);
+
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 338a02bb004d..dc523a1f270d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -49,7 +49,6 @@ nodemask_t node_online_map __read_mostly = { { [0] = 1UL } };
 EXPORT_SYMBOL(node_online_map);
 nodemask_t node_possible_map __read_mostly = NODE_MASK_ALL;
 EXPORT_SYMBOL(node_possible_map);
-struct pglist_data *pgdat_list __read_mostly;
 unsigned long totalram_pages __read_mostly;
 unsigned long totalhigh_pages __read_mostly;
 long nr_swap_pages;
@@ -1201,7 +1200,7 @@ unsigned int nr_free_highpages (void)
         pg_data_t *pgdat;
         unsigned int pages = 0;
 
-        for_each_pgdat(pgdat)
+        for_each_online_pgdat(pgdat)
                 pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
 
         return pages;
@@ -1343,7 +1342,7 @@ void get_zone_counts(unsigned long *active,
         *active = 0;
         *inactive = 0;
         *free = 0;
-        for_each_pgdat(pgdat) {
+        for_each_online_pgdat(pgdat) {
                 unsigned long l, m, n;
                 __get_zone_counts(&l, &m, &n, pgdat);
                 *active += l;
@@ -2042,7 +2041,6 @@ static __meminit void init_currently_empty_zone(struct zone *zone,
         zone_wait_table_init(zone, size);
         pgdat->nr_zones = zone_idx(zone) + 1;
 
-        zone->zone_mem_map = pfn_to_page(zone_start_pfn);
         zone->zone_start_pfn = zone_start_pfn;
 
         memmap_init(size, pgdat->node_id, zone_idx(zone), zone_start_pfn);
@@ -2170,8 +2168,9 @@ static void *frag_start(struct seq_file *m, loff_t *pos)
 {
         pg_data_t *pgdat;
         loff_t node = *pos;
-
-        for (pgdat = pgdat_list; pgdat && node; pgdat = pgdat->pgdat_next)
+        for (pgdat = first_online_pgdat();
+             pgdat && node;
+             pgdat = next_online_pgdat(pgdat))
                 --node;
 
         return pgdat;
@@ -2182,7 +2181,7 @@ static void *frag_next(struct seq_file *m, void *arg, loff_t *pos)
         pg_data_t *pgdat = (pg_data_t *)arg;
 
         (*pos)++;
-        return pgdat->pgdat_next;
+        return next_online_pgdat(pgdat);
 }
 
 static void frag_stop(struct seq_file *m, void *arg)
@@ -2483,7 +2482,7 @@ static void setup_per_zone_lowmem_reserve(void)
         struct pglist_data *pgdat;
         int j, idx;
 
-        for_each_pgdat(pgdat) {
+        for_each_online_pgdat(pgdat) {
                 for (j = 0; j < MAX_NR_ZONES; j++) {
                         struct zone *zone = pgdat->node_zones + j;
                         unsigned long present_pages = zone->present_pages;
@@ -2745,3 +2744,44 @@ void *__init alloc_large_system_hash(const char *tablename,
 
         return table;
 }
+
+#ifdef CONFIG_OUT_OF_LINE_PFN_TO_PAGE
+/*
+ * pfn <-> page translation. out-of-line version.
+ * (see asm-generic/memory_model.h)
+ */
+#if defined(CONFIG_FLATMEM)
+struct page *pfn_to_page(unsigned long pfn)
+{
+        return mem_map + (pfn - ARCH_PFN_OFFSET);
+}
+unsigned long page_to_pfn(struct page *page)
+{
+        return (page - mem_map) + ARCH_PFN_OFFSET;
+}
+#elif defined(CONFIG_DISCONTIGMEM)
+struct page *pfn_to_page(unsigned long pfn)
+{
+        int nid = arch_pfn_to_nid(pfn);
+        return NODE_DATA(nid)->node_mem_map + arch_local_page_offset(pfn,nid);
+}
+unsigned long page_to_pfn(struct page *page)
+{
+        struct pglist_data *pgdat = NODE_DATA(page_to_nid(page));
+        return (page - pgdat->node_mem_map) + pgdat->node_start_pfn;
+}
+#elif defined(CONFIG_SPARSEMEM)
+struct page *pfn_to_page(unsigned long pfn)
+{
+        return __section_mem_map_addr(__pfn_to_section(pfn)) + pfn;
+}
+
+unsigned long page_to_pfn(struct page *page)
+{
+        long section_id = page_to_section(page);
+        return page - __section_mem_map_addr(__nr_to_section(section_id));
+}
+#endif /* CONFIG_FLATMEM/DISCONTIGMME/SPARSEMEM */
+EXPORT_SYMBOL(pfn_to_page);
+EXPORT_SYMBOL(page_to_pfn);
+#endif /* CONFIG_OUT_OF_LINE_PFN_TO_PAGE */
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;
         }
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 78865c849f8f..acdf001d6941 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1305,7 +1305,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages)
 
         current->reclaim_state = &reclaim_state;
 repeat:
-        for_each_pgdat(pgdat) {
+        for_each_online_pgdat(pgdat) {
                 unsigned long freed;
 
                 freed = balance_pgdat(pgdat, nr_to_free, 0);
@@ -1335,7 +1335,7 @@ static int __devinit cpu_callback(struct notifier_block *nfb,
         cpumask_t mask;
 
         if (action == CPU_ONLINE) {
-                for_each_pgdat(pgdat) {
+                for_each_online_pgdat(pgdat) {
                         mask = node_to_cpumask(pgdat->node_id);
                         if (any_online_cpu(mask) != NR_CPUS)
                                 /* One of our CPUs online: restore mask */
@@ -1351,7 +1351,7 @@ static int __init kswapd_init(void)
         pg_data_t *pgdat;
 
         swap_setup();
-        for_each_pgdat(pgdat) {
+        for_each_online_pgdat(pgdat) {
                 pid_t pid;
 
                 pid = kernel_thread(kswapd, pgdat, CLONE_KERNEL);