[PATCH] x86_64: add x86-64 support for memory hot-add

Add x86-64 specific memory hot-add functions, Kconfig options,
and runtime kernel page table update functions to make
hot-add usable on x86-64 machines.  Also, fixup the nefarious
conditional locking and exports pointed out by Andi.

Tested on Intel and IBM x86-64 memory hot-add capable systems.

Signed-off-by: Matt Tolentino <matthew.e.tolentino@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

2 files changed, 134 insertions, 33 deletions

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 787feff0e8a5..2f9deca31cc9 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -305,7 +305,11 @@ config ARCH_DISCONTIGMEM_DEFAULT
 
 config ARCH_SPARSEMEM_ENABLE
         def_bool y
-        depends on NUMA
+        depends on (NUMA || EXPERIMENTAL)
+
+config ARCH_MEMORY_PROBE
+        def_bool y
+        depends on MEMORY_HOTPLUG
 
 config ARCH_FLATMEM_ENABLE
         def_bool y
@@ -315,6 +319,7 @@ source "mm/Kconfig"
 
 config HAVE_ARCH_EARLY_PFN_TO_NID
         def_bool y
+        depends on NUMA
 
 config NR_CPUS
         int "Maximum number of CPUs (2-256)"
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 307d1ac0ef33..7af1742aa958 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -24,6 +24,8 @@
 #include <linux/proc_fs.h>
 #include <linux/pci.h>
 #include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/memory_hotplug.h>
 
 #include <asm/processor.h>
 #include <asm/system.h>
@@ -180,13 +182,19 @@ static  struct temp_map {
         {}
 }; 
 
-static __init void *alloc_low_page(int *index, unsigned long *phys) 
+static __meminit void *alloc_low_page(int *index, unsigned long *phys)
 { 
         struct temp_map *ti;
         int i; 
         unsigned long pfn = table_end++, paddr; 
         void *adr;
 
+        if (after_bootmem) {
+                adr = (void *)get_zeroed_page(GFP_ATOMIC);
+                *phys = __pa(adr);
+                return adr;
+        }
+
         if (pfn >= end_pfn) 
                 panic("alloc_low_page: ran out of memory"); 
         for (i = 0; temp_mappings[i].allocated; i++) {
@@ -199,55 +207,86 @@ static __init void *alloc_low_page(int *index, unsigned long *phys)
         ti->allocated = 1; 
         __flush_tlb();         
         adr = ti->address + ((pfn << PAGE_SHIFT) & ~PMD_MASK); 
+        memset(adr, 0, PAGE_SIZE);
         *index = i; 
         *phys  = pfn * PAGE_SIZE;  
         return adr; 
 } 
 
-static __init void unmap_low_page(int i)
+static __meminit void unmap_low_page(int i)
 { 
-        struct temp_map *ti = &temp_mappings[i];
+        struct temp_map *ti;
+
+        if (after_bootmem)
+                return;
+
+        ti = &temp_mappings[i];
         set_pmd(ti->pmd, __pmd(0));
         ti->allocated = 0; 
 } 
 
-static void __init phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
+static void __meminit
+phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end)
+{
+        int i;
+
+        for (i = 0; i < PTRS_PER_PMD; pmd++, i++, address += PMD_SIZE) {
+                unsigned long entry;
+
+                if (address > end) {
+                        for (; i < PTRS_PER_PMD; i++, pmd++)
+                                set_pmd(pmd, __pmd(0));
+                        break;
+                }
+                entry = _PAGE_NX|_PAGE_PSE|_KERNPG_TABLE|_PAGE_GLOBAL|address;
+                entry &= __supported_pte_mask;
+                set_pmd(pmd, __pmd(entry));
+        }
+}
+
+static void __meminit
+phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
+{
+        pmd_t *pmd = pmd_offset(pud, (unsigned long)__va(address));
+
+        if (pmd_none(*pmd)) {
+                spin_lock(&init_mm.page_table_lock);
+                phys_pmd_init(pmd, address, end);
+                spin_unlock(&init_mm.page_table_lock);
+                __flush_tlb_all();
+        }
+}
+
+static void __meminit phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
 { 
-        long i, j; 
+        long i = pud_index(address);
 
-        i = pud_index(address);
         pud = pud + i;
+
+        if (after_bootmem && pud_val(*pud)) {
+                phys_pmd_update(pud, address, end);
+                return;
+        }
+
         for (; i < PTRS_PER_PUD; pud++, i++) {
                 int map; 
                 unsigned long paddr, pmd_phys;
                 pmd_t *pmd;
 
-                paddr = address + i*PUD_SIZE;
-                if (paddr >= end) { 
-                        for (; i < PTRS_PER_PUD; i++, pud++) 
-                                set_pud(pud, __pud(0)); 
+                paddr = (address & PGDIR_MASK) + i*PUD_SIZE;
+                if (paddr >= end)
                         break;
-                } 
 
-                if (!e820_mapped(paddr, paddr+PUD_SIZE, 0)) { 
+                if (!after_bootmem && !e820_mapped(paddr, paddr+PUD_SIZE, 0)) {
                         set_pud(pud, __pud(0)); 
                         continue;
                 } 
 
                 pmd = alloc_low_page(&map, &pmd_phys);
+                spin_lock(&init_mm.page_table_lock);
                 set_pud(pud, __pud(pmd_phys | _KERNPG_TABLE));
-                for (j = 0; j < PTRS_PER_PMD; pmd++, j++, paddr += PMD_SIZE) {
-                        unsigned long pe;
-
-                        if (paddr >= end) { 
-                                for (; j < PTRS_PER_PMD; j++, pmd++)
-                                        set_pmd(pmd,  __pmd(0)); 
-                                break;
-                }
-                        pe = _PAGE_NX|_PAGE_PSE | _KERNPG_TABLE | _PAGE_GLOBAL | paddr;
-                        pe &= __supported_pte_mask;
-                        set_pmd(pmd, __pmd(pe));
-                }
+                phys_pmd_init(pmd, paddr, end);
+                spin_unlock(&init_mm.page_table_lock);
                 unmap_low_page(map);
         }
         __flush_tlb();
@@ -272,12 +311,15 @@ static void __init find_early_table_space(unsigned long end)
 
         table_start >>= PAGE_SHIFT;
         table_end = table_start;
+
+        early_printk("kernel direct mapping tables up to %lx @ %lx-%lx\n",
+                end, table_start << PAGE_SHIFT, table_end << PAGE_SHIFT);
 }
 
 /* Setup the direct mapping of the physical memory at PAGE_OFFSET.
    This runs before bootmem is initialized and gets pages directly from the 
    physical memory. To access them they are temporarily mapped. */
-void __init init_memory_mapping(unsigned long start, unsigned long end)
+void __meminit init_memory_mapping(unsigned long start, unsigned long end)
 { 
         unsigned long next; 
 
@@ -289,7 +331,8 @@ void __init init_memory_mapping(unsigned long start, unsigned long end)
          * mapped.  Unfortunately this is done currently before the nodes are 
          * discovered.
          */
-        find_early_table_space(end);
+        if (!after_bootmem)
+                find_early_table_space(end);
 
         start = (unsigned long)__va(start);
         end = (unsigned long)__va(end);
@@ -297,20 +340,26 @@ void __init init_memory_mapping(unsigned long start, unsigned long end)
         for (; start < end; start = next) {
                 int map;
                 unsigned long pud_phys; 
-                pud_t *pud = alloc_low_page(&map, &pud_phys);
+                pgd_t *pgd = pgd_offset_k(start);
+                pud_t *pud;
+
+                if (after_bootmem)
+                        pud = pud_offset_k(pgd, __PAGE_OFFSET);
+                else
+                        pud = alloc_low_page(&map, &pud_phys);
+
                 next = start + PGDIR_SIZE;
                 if (next > end) 
                         next = end; 
                 phys_pud_init(pud, __pa(start), __pa(next));
-                set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
+                if (!after_bootmem)
+                        set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
                 unmap_low_page(map);   
         } 
 
-        asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features));
+        if (!after_bootmem)
+                asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features));
         __flush_tlb_all();
-        early_printk("kernel direct mapping tables upto %lx @ %lx-%lx\n", end, 
-               table_start<<PAGE_SHIFT, 
-               table_end<<PAGE_SHIFT);
 }
 
 void __cpuinit zap_low_mappings(int cpu)
@@ -385,6 +434,9 @@ size_zones(unsigned long *z, unsigned long *h,
 void __init paging_init(void)
 {
         unsigned long zones[MAX_NR_ZONES], holes[MAX_NR_ZONES];
+
+        memory_present(0, 0, end_pfn);
+        sparse_init();
         size_zones(zones, holes, 0, end_pfn);
         free_area_init_node(0, NODE_DATA(0), zones,
                             __pa(PAGE_OFFSET) >> PAGE_SHIFT, holes);
@@ -425,6 +477,50 @@ void __init clear_kernel_mapping(unsigned long address, unsigned long size)
         __flush_tlb_all();
 } 
 
+/*
+ * Memory hotplug specific functions
+ * These are only for non-NUMA machines right now.
+ */
+#ifdef CONFIG_MEMORY_HOTPLUG
+
+void online_page(struct page *page)
+{
+        ClearPageReserved(page);
+        set_page_count(page, 1);
+        __free_page(page);
+        totalram_pages++;
+        num_physpages++;
+}
+
+int add_memory(u64 start, u64 size)
+{
+        struct pglist_data *pgdat = NODE_DATA(0);
+        struct zone *zone = pgdat->node_zones + MAX_NR_ZONES-2;
+        unsigned long start_pfn = start >> PAGE_SHIFT;
+        unsigned long nr_pages = size >> PAGE_SHIFT;
+        int ret;
+
+        ret = __add_pages(zone, start_pfn, nr_pages);
+        if (ret)
+                goto error;
+
+        init_memory_mapping(start, (start + size -1));
+
+        return ret;
+error:
+        printk("%s: Problem encountered in __add_pages!\n", __func__);
+        return ret;
+}
+EXPORT_SYMBOL_GPL(add_memory);
+
+int remove_memory(u64 start, u64 size)
+{
+        return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(remove_memory);
+
+#endif
+
 static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules,
                          kcore_vsyscall;