1 files changed, 45 insertions, 23 deletions

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 5ed471b58f4f..9f22090df7dd 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -12,6 +12,7 @@
 #include <linux/threads.h>
 #include <linux/numa.h>
 #include <linux/init.h>
+#include <linux/seqlock.h>
 #include <asm/atomic.h>
 
 /* Free memory management - zoned buddy allocator.  */
@@ -70,10 +71,11 @@ struct per_cpu_pageset {
 #endif
 
 #define ZONE_DMA                0
-#define ZONE_NORMAL             1
-#define ZONE_HIGHMEM            2
+#define ZONE_DMA32              1
+#define ZONE_NORMAL             2
+#define ZONE_HIGHMEM            3
 
-#define MAX_NR_ZONES            3       /* Sync this with ZONES_SHIFT */
+#define MAX_NR_ZONES            4       /* Sync this with ZONES_SHIFT */
 #define ZONES_SHIFT             2       /* ceil(log2(MAX_NR_ZONES)) */
 
 
@@ -89,27 +91,18 @@ struct per_cpu_pageset {
  * will be a maximum of 4 (2 ** 2) zonelists, for 3 modifiers there will
  * be 8 (2 ** 3) zonelists.  GFP_ZONETYPES defines the number of possible
  * combinations of zone modifiers in "zone modifier space".
+ *
+ * NOTE! Make sure this matches the zones in <linux/gfp.h>
  */
-#define GFP_ZONEMASK    0x03
-/*
- * As an optimisation any zone modifier bits which are only valid when
- * no other zone modifier bits are set (loners) should be placed in
- * the highest order bits of this field.  This allows us to reduce the
- * extent of the zonelists thus saving space.  For example in the case
- * of three zone modifier bits, we could require up to eight zonelists.
- * If the left most zone modifier is a "loner" then the highest valid
- * zonelist would be four allowing us to allocate only five zonelists.
- * Use the first form when the left most bit is not a "loner", otherwise
- * use the second.
- */
-/* #define GFP_ZONETYPES        (GFP_ZONEMASK + 1) */           /* Non-loner */
-#define GFP_ZONETYPES   ((GFP_ZONEMASK + 1) / 2 + 1)            /* Loner */
+#define GFP_ZONEMASK    0x07
+#define GFP_ZONETYPES   5
 
 /*
  * On machines where it is needed (eg PCs) we divide physical memory
- * into multiple physical zones. On a PC we have 3 zones:
+ * into multiple physical zones. On a PC we have 4 zones:
  *
  * ZONE_DMA       < 16 MB       ISA DMA capable memory
+ * ZONE_DMA32        0 MB       Empty
  * ZONE_NORMAL  16-896 MB       direct mapped by the kernel
  * ZONE_HIGHMEM  > 896 MB       only page cache and user processes
  */
@@ -137,6 +130,10 @@ struct zone {
          * free areas of different sizes
          */
         spinlock_t              lock;
+#ifdef CONFIG_MEMORY_HOTPLUG
+        /* see spanned/present_pages for more description */
+        seqlock_t               span_seqlock;
+#endif
         struct free_area        free_area[MAX_ORDER];
 
 
@@ -220,6 +217,16 @@ struct zone {
         /* zone_start_pfn == zone_start_paddr >> PAGE_SHIFT */
         unsigned long           zone_start_pfn;
 
+        /*
+         * zone_start_pfn, spanned_pages and present_pages are all
+         * protected by span_seqlock.  It is a seqlock because it has
+         * to be read outside of zone->lock, and it is done in the main
+         * allocator path.  But, it is written quite infrequently.
+         *
+         * The lock is declared along with zone->lock because it is
+         * frequently read in proximity to zone->lock.  It's good to
+         * give them a chance of being in the same cacheline.
+         */
         unsigned long           spanned_pages;  /* total size, including holes */
         unsigned long           present_pages;  /* amount of memory (excluding holes) */
 
@@ -273,6 +280,16 @@ typedef struct pglist_data {
         struct page *node_mem_map;
 #endif
         struct bootmem_data *bdata;
+#ifdef CONFIG_MEMORY_HOTPLUG
+        /*
+         * Must be held any time you expect node_start_pfn, node_present_pages
+         * or node_spanned_pages stay constant.  Holding this will also
+         * guarantee that any pfn_valid() stays that way.
+         *
+         * Nests above zone->lock and zone->size_seqlock.
+         */
+        spinlock_t node_size_lock;
+#endif
         unsigned long node_start_pfn;
         unsigned long node_present_pages; /* total number of physical pages */
         unsigned long node_spanned_pages; /* total size of physical page
@@ -293,6 +310,8 @@ typedef struct pglist_data {
 #endif
 #define nid_page_nr(nid, pagenr)        pgdat_page_nr(NODE_DATA(nid),(pagenr))
 
+#include <linux/memory_hotplug.h>
+
 extern struct pglist_data *pgdat_list;
 
 void __get_zone_counts(unsigned long *active, unsigned long *inactive,
@@ -302,7 +321,7 @@ void get_zone_counts(unsigned long *active, unsigned long *inactive,
 void build_all_zonelists(void);
 void wakeup_kswapd(struct zone *zone, int order);
 int zone_watermark_ok(struct zone *z, int order, unsigned long mark,
-                int alloc_type, int can_try_harder, int gfp_high);
+                int classzone_idx, int alloc_flags);
 
 #ifdef CONFIG_HAVE_MEMORY_PRESENT
 void memory_present(int nid, unsigned long start, unsigned long end);
@@ -406,7 +425,9 @@ int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int, struct file *,
 
 #include <linux/topology.h>
 /* Returns the number of the current Node. */
+#ifndef numa_node_id
 #define numa_node_id()          (cpu_to_node(raw_smp_processor_id()))
+#endif
 
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 
@@ -426,12 +447,12 @@ extern struct pglist_data contig_page_data;
 #include <asm/sparsemem.h>
 #endif
 
-#if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED)
+#if BITS_PER_LONG == 32
 /*
- * with 32 bit page->flags field, we reserve 8 bits for node/zone info.
- * there are 3 zones (2 bits) and this leaves 8-2=6 bits for nodes.
+ * with 32 bit page->flags field, we reserve 9 bits for node/zone info.
+ * there are 4 zones (3 bits) and this leaves 9-3=6 bits for nodes.
  */
-#define FLAGS_RESERVED          8
+#define FLAGS_RESERVED          9
 
 #elif BITS_PER_LONG == 64
 /*
@@ -509,6 +530,7 @@ static inline struct mem_section *__nr_to_section(unsigned long nr)
                 return NULL;
         return &mem_section[SECTION_NR_TO_ROOT(nr)][nr & SECTION_ROOT_MASK];
 }
+extern int __section_nr(struct mem_section* ms);
 
 /*
  * We use the lower bits of the mem_map pointer to store