4 files changed, 143 insertions, 11 deletions

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index bd33be24a386..79190891fbc5 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -87,6 +87,10 @@ void __init setup_per_cpu_areas(void)
         int i;
         unsigned long size;
 
+#ifdef CONFIG_HOTPLUG_CPU
+        prefill_possible_map();
+#endif
+
         /* Copy section for each CPU (we discard the original) */
         size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
 #ifdef CONFIG_MODULES
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index e12d7baeb33e..658a81b33f3b 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -892,7 +892,7 @@ static __init void disable_smp(void)
  * those NR_CPUS, hence cpu_possible_map represents entire NR_CPUS range.
  * - Ashok Raj
  */
-static void prefill_possible_map(void)
+__init void prefill_possible_map(void)
 {
         int i;
         for (i = 0; i < NR_CPUS; i++)
@@ -967,10 +967,6 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
         current_cpu_data = boot_cpu_data;
         current_thread_info()->cpu = 0;  /* needed? */
 
-#ifdef CONFIG_HOTPLUG_CPU
-        prefill_possible_map();
-#endif
-
         if (smp_sanity_check(max_cpus) < 0) {
                 printk(KERN_INFO "SMP disabled\n");
                 disable_smp();
diff --git a/arch/x86_64/kernel/suspend.c b/arch/x86_64/kernel/suspend.c
index ebb9abf3ce6d..f066c6ab3618 100644
--- a/arch/x86_64/kernel/suspend.c
+++ b/arch/x86_64/kernel/suspend.c
@@ -11,6 +11,8 @@
 #include <linux/smp.h>
 #include <linux/suspend.h>
 #include <asm/proto.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
 
 struct saved_context saved_context;
 
@@ -140,4 +142,129 @@ void fix_processor_context(void)
 
 }
 
+#ifdef CONFIG_SOFTWARE_SUSPEND
+/* Defined in arch/x86_64/kernel/suspend_asm.S */
+extern int restore_image(void);
 
+pgd_t *temp_level4_pgt;
+
+static void **pages;
+
+static inline void *__add_page(void)
+{
+        void **c;
+
+        c = (void **)get_usable_page(GFP_ATOMIC);
+        if (c) {
+                *c = pages;
+                pages = c;
+        }
+        return c;
+}
+
+static inline void *__next_page(void)
+{
+        void **c;
+
+        c = pages;
+        if (c) {
+                pages = *c;
+                *c = NULL;
+        }
+        return c;
+}
+
+/*
+ * Try to allocate as many usable pages as needed and daisy chain them.
+ * If one allocation fails, free the pages allocated so far
+ */
+static int alloc_usable_pages(unsigned long n)
+{
+        void *p;
+
+        pages = NULL;
+        do
+                if (!__add_page())
+                        break;
+        while (--n);
+        if (n) {
+                p = __next_page();
+                while (p) {
+                        free_page((unsigned long)p);
+                        p = __next_page();
+                }
+                return -ENOMEM;
+        }
+        return 0;
+}
+
+static void res_phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
+{
+        long i, j;
+
+        i = pud_index(address);
+        pud = pud + i;
+        for (; i < PTRS_PER_PUD; pud++, i++) {
+                unsigned long paddr;
+                pmd_t *pmd;
+
+                paddr = address + i*PUD_SIZE;
+                if (paddr >= end)
+                        break;
+
+                pmd = (pmd_t *)__next_page();
+                set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE));
+                for (j = 0; j < PTRS_PER_PMD; pmd++, j++, paddr += PMD_SIZE) {
+                        unsigned long pe;
+
+                        if (paddr >= end)
+                                break;
+                        pe = _PAGE_NX | _PAGE_PSE | _KERNPG_TABLE | paddr;
+                        pe &= __supported_pte_mask;
+                        set_pmd(pmd, __pmd(pe));
+                }
+        }
+}
+
+static void set_up_temporary_mappings(void)
+{
+        unsigned long start, end, next;
+
+        temp_level4_pgt = (pgd_t *)__next_page();
+
+        /* It is safe to reuse the original kernel mapping */
+        set_pgd(temp_level4_pgt + pgd_index(__START_KERNEL_map),
+                init_level4_pgt[pgd_index(__START_KERNEL_map)]);
+
+        /* Set up the direct mapping from scratch */
+        start = (unsigned long)pfn_to_kaddr(0);
+        end = (unsigned long)pfn_to_kaddr(end_pfn);
+
+        for (; start < end; start = next) {
+                pud_t *pud = (pud_t *)__next_page();
+                next = start + PGDIR_SIZE;
+                if (next > end)
+                        next = end;
+                res_phys_pud_init(pud, __pa(start), __pa(next));
+                set_pgd(temp_level4_pgt + pgd_index(start),
+                        mk_kernel_pgd(__pa(pud)));
+        }
+}
+
+int swsusp_arch_resume(void)
+{
+        unsigned long n;
+
+        n = ((end_pfn << PAGE_SHIFT) + PUD_SIZE - 1) >> PUD_SHIFT;
+        n += (n + PTRS_PER_PUD - 1) / PTRS_PER_PUD + 1;
+        pr_debug("swsusp_arch_resume(): pages needed = %lu\n", n);
+        if (alloc_usable_pages(n)) {
+                free_eaten_memory();
+                return -ENOMEM;
+        }
+        /* We have got enough memory and from now on we cannot recover */
+        set_up_temporary_mappings();
+        restore_image();
+        return 0;
+}
+#endif /* CONFIG_SOFTWARE_SUSPEND */
diff --git a/arch/x86_64/kernel/suspend_asm.S b/arch/x86_64/kernel/suspend_asm.S
index 4d659e97df10..320b6fb00cca 100644
--- a/arch/x86_64/kernel/suspend_asm.S
+++ b/arch/x86_64/kernel/suspend_asm.S
@@ -39,12 +39,13 @@ ENTRY(swsusp_arch_suspend)
         call swsusp_save
         ret
 
-ENTRY(swsusp_arch_resume)
-        /* set up cr3 */        
-        leaq    init_level4_pgt(%rip),%rax
-        subq    $__START_KERNEL_map,%rax
-        movq    %rax,%cr3
-
+ENTRY(restore_image)
+        /* switch to temporary page tables */
+        movq    $__PAGE_OFFSET, %rdx
+        movq    temp_level4_pgt(%rip), %rax
+        subq    %rdx, %rax
+        movq    %rax, %cr3
+        /* Flush TLB */
         movq    mmu_cr4_features(%rip), %rax
         movq    %rax, %rdx
         andq    $~(1<<7), %rdx  # PGE
@@ -69,6 +70,10 @@ loop:
         movq    pbe_next(%rdx), %rdx
         jmp     loop
 done:
+        /* go back to the original page tables */
+        leaq    init_level4_pgt(%rip), %rax
+        subq    $__START_KERNEL_map, %rax
+        movq    %rax, %cr3
         /* Flush TLB, including "global" things (vmalloc) */
         movq    mmu_cr4_features(%rip), %rax
         movq    %rax, %rdx