netns xfrm: fix "ip xfrm state|policy count" misreport

[safe/jmp/linux-2.6] / mm / sparse.c

diff --git a/mm/sparse.c b/mm/sparse.c
index 7a36509..6ce4aab 100644 (file)
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -12,7 +12,6 @@
 #include <asm/dma.h>
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
-#include "internal.h"
 
 /*
  * Permanent SPARSEMEM data:
@@ -63,9 +62,12 @@ static struct mem_section noinline __init_refok *sparse_index_alloc(int nid)
        unsigned long array_size = SECTIONS_PER_ROOT *
                                   sizeof(struct mem_section);
 
-       if (slab_is_available())
-               section = kmalloc_node(array_size, GFP_KERNEL, nid);
-       else
+       if (slab_is_available()) {
+               if (node_state(nid, N_HIGH_MEMORY))
+                       section = kmalloc_node(array_size, GFP_KERNEL, nid);
+               else
+                       section = kmalloc(array_size, GFP_KERNEL);
+       } else
                section = alloc_bootmem_node(NODE_DATA(nid), array_size);
 
        if (section)
@@ -165,9 +167,7 @@ void __meminit mminit_validate_memmodel_limits(unsigned long *start_pfn,
                WARN_ON_ONCE(1);
                *start_pfn = max_sparsemem_pfn;
                *end_pfn = max_sparsemem_pfn;
-       }
-
-       if (*end_pfn > max_sparsemem_pfn) {
+       } else if (*end_pfn > max_sparsemem_pfn) {
                mminit_dprintk(MMINIT_WARNING, "pfnvalidation",
                        "End of range %lu -> %lu exceeds SPARSEMEM max %lu\n",
                        *start_pfn, *end_pfn, max_sparsemem_pfn);
@@ -269,16 +269,92 @@ static unsigned long *__kmalloc_section_usemap(void)
 }
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
+#ifdef CONFIG_MEMORY_HOTREMOVE
+static unsigned long * __init
+sparse_early_usemap_alloc_pgdat_section(struct pglist_data *pgdat)
+{
+       unsigned long section_nr;
+
+       /*
+        * A page may contain usemaps for other sections preventing the
+        * page being freed and making a section unremovable while
+        * other sections referencing the usemap retmain active. Similarly,
+        * a pgdat can prevent a section being removed. If section A
+        * contains a pgdat and section B contains the usemap, both
+        * sections become inter-dependent. This allocates usemaps
+        * from the same section as the pgdat where possible to avoid
+        * this problem.
+        */
+       section_nr = pfn_to_section_nr(__pa(pgdat) >> PAGE_SHIFT);
+       return alloc_bootmem_section(usemap_size(), section_nr);
+}
+
+static void __init check_usemap_section_nr(int nid, unsigned long *usemap)
+{
+       unsigned long usemap_snr, pgdat_snr;
+       static unsigned long old_usemap_snr = NR_MEM_SECTIONS;
+       static unsigned long old_pgdat_snr = NR_MEM_SECTIONS;
+       struct pglist_data *pgdat = NODE_DATA(nid);
+       int usemap_nid;
+
+       usemap_snr = pfn_to_section_nr(__pa(usemap) >> PAGE_SHIFT);
+       pgdat_snr = pfn_to_section_nr(__pa(pgdat) >> PAGE_SHIFT);
+       if (usemap_snr == pgdat_snr)
+               return;
+
+       if (old_usemap_snr == usemap_snr && old_pgdat_snr == pgdat_snr)
+               /* skip redundant message */
+               return;
+
+       old_usemap_snr = usemap_snr;
+       old_pgdat_snr = pgdat_snr;
+
+       usemap_nid = sparse_early_nid(__nr_to_section(usemap_snr));
+       if (usemap_nid != nid) {
+               printk(KERN_INFO
+                      "node %d must be removed before remove section %ld\n",
+                      nid, usemap_snr);
+               return;
+       }
+       /*
+        * There is a circular dependency.
+        * Some platforms allow un-removable section because they will just
+        * gather other removable sections for dynamic partitioning.
+        * Just notify un-removable section's number here.
+        */
+       printk(KERN_INFO "Section %ld and %ld (node %d)", usemap_snr,
+              pgdat_snr, nid);
+       printk(KERN_CONT
+              " have a circular dependency on usemap and pgdat allocations\n");
+}
+#else
+static unsigned long * __init
+sparse_early_usemap_alloc_pgdat_section(struct pglist_data *pgdat)
+{
+       return NULL;
+}
+
+static void __init check_usemap_section_nr(int nid, unsigned long *usemap)
+{
+}
+#endif /* CONFIG_MEMORY_HOTREMOVE */
+
 static unsigned long *__init sparse_early_usemap_alloc(unsigned long pnum)
 {
        unsigned long *usemap;
        struct mem_section *ms = __nr_to_section(pnum);
        int nid = sparse_early_nid(ms);
 
-       usemap = alloc_bootmem_node(NODE_DATA(nid), usemap_size());
+       usemap = sparse_early_usemap_alloc_pgdat_section(NODE_DATA(nid));
        if (usemap)
                return usemap;
 
+       usemap = alloc_bootmem_node(NODE_DATA(nid), usemap_size());
+       if (usemap) {
+               check_usemap_section_nr(nid, usemap);
+               return usemap;
+       }
+
        /* Stupid: suppress gcc warning for SPARSEMEM && !NUMA */
        nid = 0;
 
@@ -301,7 +377,7 @@ struct page __init *sparse_mem_map_populate(unsigned long pnum, int nid)
 }
 #endif /* !CONFIG_SPARSEMEM_VMEMMAP */
 
-struct page __init *sparse_early_mem_map_alloc(unsigned long pnum)
+static struct page __init *sparse_early_mem_map_alloc(unsigned long pnum)
 {
        struct page *map;
        struct mem_section *ms = __nr_to_section(pnum);
@@ -495,7 +571,7 @@ static void free_section_usemap(struct page *memmap, unsigned long *usemap)
  * set.  If this is <=0, then that means that the passed-in
  * map was not consumed and must be freed.
  */
-int sparse_add_one_section(struct zone *zone, unsigned long start_pfn,
+int __meminit sparse_add_one_section(struct zone *zone, unsigned long start_pfn,
                           int nr_pages)
 {
        unsigned long section_nr = pfn_to_section_nr(start_pfn);