6 files changed, 72 insertions, 8 deletions

diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c
index e55fd05da0f5..7329ec9fcc99 100644
--- a/arch/i386/kernel/paravirt.c
+++ b/arch/i386/kernel/paravirt.c
@@ -550,6 +550,12 @@ struct paravirt_ops paravirt_ops = {
         .flush_tlb_kernel = native_flush_tlb_global,
         .flush_tlb_single = native_flush_tlb_single,
 
+        .alloc_pt = (void *)native_nop,
+        .alloc_pd = (void *)native_nop,
+        .alloc_pd_clone = (void *)native_nop,
+        .release_pt = (void *)native_nop,
+        .release_pd = (void *)native_nop,
+
         .set_pte = native_set_pte,
         .set_pte_at = native_set_pte_at,
         .set_pmd = native_set_pmd,
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index c5c5ea700cc7..ae436882af7a 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -62,6 +62,7 @@ static pmd_t * __init one_md_table_init(pgd_t *pgd)
                 
 #ifdef CONFIG_X86_PAE
         pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+        paravirt_alloc_pd(__pa(pmd_table) >> PAGE_SHIFT);
         set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
         pud = pud_offset(pgd, 0);
         if (pmd_table != pmd_offset(pud, 0)) 
@@ -82,6 +83,7 @@ static pte_t * __init one_page_table_init(pmd_t *pmd)
 {
         if (pmd_none(*pmd)) {
                 pte_t *page_table = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+                paravirt_alloc_pt(__pa(page_table) >> PAGE_SHIFT);
                 set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE));
                 if (page_table != pte_offset_kernel(pmd, 0))
                         BUG();  
@@ -345,6 +347,8 @@ static void __init pagetable_init (void)
         /* Init entries of the first-level page table to the zero page */
         for (i = 0; i < PTRS_PER_PGD; i++)
                 set_pgd(pgd_base + i, __pgd(__pa(empty_zero_page) | _PAGE_PRESENT));
+#else
+        paravirt_alloc_pd(__pa(swapper_pg_dir) >> PAGE_SHIFT);
 #endif
 
         /* Enable PSE if available */
diff --git a/arch/i386/mm/pageattr.c b/arch/i386/mm/pageattr.c
index e223b1d4981c..412ebbd8adb0 100644
--- a/arch/i386/mm/pageattr.c
+++ b/arch/i386/mm/pageattr.c
@@ -60,6 +60,7 @@ static struct page *split_large_page(unsigned long address, pgprot_t prot,
         address = __pa(address);
         addr = address & LARGE_PAGE_MASK; 
         pbase = (pte_t *)page_address(base);
+        paravirt_alloc_pt(page_to_pfn(base));
         for (i = 0; i < PTRS_PER_PTE; i++, addr += PAGE_SIZE) {
                set_pte(&pbase[i], pfn_pte(addr >> PAGE_SHIFT,
                                           addr == address ? prot : ref_prot));
@@ -172,6 +173,7 @@ __change_page_attr(struct page *page, pgprot_t prot)
         if (!PageReserved(kpte_page)) {
                 if (cpu_has_pse && (page_private(kpte_page) == 0)) {
                         ClearPagePrivate(kpte_page);
+                        paravirt_release_pt(page_to_pfn(kpte_page));
                         list_add(&kpte_page->lru, &df_list);
                         revert_page(kpte_page, address);
                 }
diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index f349eaf450b0..b5f538f52272 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -248,9 +248,15 @@ void pgd_ctor(void *pgd, struct kmem_cache *cache, unsigned long unused)
         clone_pgd_range((pgd_t *)pgd + USER_PTRS_PER_PGD,
                         swapper_pg_dir + USER_PTRS_PER_PGD,
                         KERNEL_PGD_PTRS);
+
         if (PTRS_PER_PMD > 1)
                 return;
 
+        /* must happen under lock */
+        paravirt_alloc_pd_clone(__pa(pgd) >> PAGE_SHIFT,
+                        __pa(swapper_pg_dir) >> PAGE_SHIFT,
+                        USER_PTRS_PER_PGD, PTRS_PER_PGD - USER_PTRS_PER_PGD);
+
         pgd_list_add(pgd);
         spin_unlock_irqrestore(&pgd_lock, flags);
 }
@@ -260,6 +266,7 @@ void pgd_dtor(void *pgd, struct kmem_cache *cache, unsigned long unused)
 {
         unsigned long flags; /* can be called from interrupt context */
 
+        paravirt_release_pd(__pa(pgd) >> PAGE_SHIFT);
         spin_lock_irqsave(&pgd_lock, flags);
         pgd_list_del(pgd);
         spin_unlock_irqrestore(&pgd_lock, flags);
@@ -277,13 +284,18 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
                 pmd_t *pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
                 if (!pmd)
                         goto out_oom;
+                paravirt_alloc_pd(__pa(pmd) >> PAGE_SHIFT);
                 set_pgd(&pgd[i], __pgd(1 + __pa(pmd)));
         }
         return pgd;
 
 out_oom:
-        for (i--; i >= 0; i--)
-                kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
+        for (i--; i >= 0; i--) {
+                pgd_t pgdent = pgd[i];
+                void* pmd = (void *)__va(pgd_val(pgdent)-1);
+                paravirt_release_pd(__pa(pmd) >> PAGE_SHIFT);
+                kmem_cache_free(pmd_cache, pmd);
+        }
         kmem_cache_free(pgd_cache, pgd);
         return NULL;
 }
@@ -294,8 +306,12 @@ void pgd_free(pgd_t *pgd)
 
         /* in the PAE case user pgd entries are overwritten before usage */
         if (PTRS_PER_PMD > 1)
-                for (i = 0; i < USER_PTRS_PER_PGD; ++i)
-                        kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
+                for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
+                        pgd_t pgdent = pgd[i];
+                        void* pmd = (void *)__va(pgd_val(pgdent)-1);
+                        paravirt_release_pd(__pa(pmd) >> PAGE_SHIFT);
+                        kmem_cache_free(pmd_cache, pmd);
+                }
         /* in the non-PAE case, free_pgtables() clears user pgd entries */
         kmem_cache_free(pgd_cache, pgd);
 }
diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h
index 9f06265065f4..53da276a2ec2 100644
--- a/include/asm-i386/paravirt.h
+++ b/include/asm-i386/paravirt.h
@@ -127,6 +127,12 @@ struct paravirt_ops
         void (fastcall *flush_tlb_kernel)(void);
         void (fastcall *flush_tlb_single)(u32 addr);
 
+        void (fastcall *alloc_pt)(u32 pfn);
+        void (fastcall *alloc_pd)(u32 pfn);
+        void (fastcall *alloc_pd_clone)(u32 pfn, u32 clonepfn, u32 start, u32 count);
+        void (fastcall *release_pt)(u32 pfn);
+        void (fastcall *release_pd)(u32 pfn);
+
         void (fastcall *set_pte)(pte_t *ptep, pte_t pteval);
         void (fastcall *set_pte_at)(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval);
         void (fastcall *set_pmd)(pmd_t *pmdp, pmd_t pmdval);
@@ -320,6 +326,14 @@ static inline unsigned long apic_read(unsigned long reg)
 #define __flush_tlb_global() paravirt_ops.flush_tlb_kernel()
 #define __flush_tlb_single(addr) paravirt_ops.flush_tlb_single(addr)
 
+#define paravirt_alloc_pt(pfn) paravirt_ops.alloc_pt(pfn)
+#define paravirt_release_pt(pfn) paravirt_ops.release_pt(pfn)
+
+#define paravirt_alloc_pd(pfn) paravirt_ops.alloc_pd(pfn)
+#define paravirt_alloc_pd_clone(pfn, clonepfn, start, count) \
+        paravirt_ops.alloc_pd_clone(pfn, clonepfn, start, count)
+#define paravirt_release_pd(pfn) paravirt_ops.release_pd(pfn)
+
 static inline void set_pte(pte_t *ptep, pte_t pteval)
 {
         paravirt_ops.set_pte(ptep, pteval);
diff --git a/include/asm-i386/pgalloc.h b/include/asm-i386/pgalloc.h
index 4b1e61359f89..c8dc2d0141a7 100644
--- a/include/asm-i386/pgalloc.h
+++ b/include/asm-i386/pgalloc.h
@@ -5,13 +5,31 @@
 #include <linux/threads.h>
 #include <linux/mm.h>           /* for struct page */
 
-#define pmd_populate_kernel(mm, pmd, pte) \
-                set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)))
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define paravirt_alloc_pt(pfn) do { } while (0)
+#define paravirt_alloc_pd(pfn) do { } while (0)
+#define paravirt_alloc_pd(pfn) do { } while (0)
+#define paravirt_alloc_pd_clone(pfn, clonepfn, start, count) do { } while (0)
+#define paravirt_release_pt(pfn) do { } while (0)
+#define paravirt_release_pd(pfn) do { } while (0)
+#endif
+
+#define pmd_populate_kernel(mm, pmd, pte)                       \
+do {                                                            \
+        paravirt_alloc_pt(__pa(pte) >> PAGE_SHIFT);             \
+        set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)));           \
+} while (0)
 
 #define pmd_populate(mm, pmd, pte)                              \
+do {                                                            \
+        paravirt_alloc_pt(page_to_pfn(pte));                    \
         set_pmd(pmd, __pmd(_PAGE_TABLE +                        \
                 ((unsigned long long)page_to_pfn(pte) <<        \
-                        (unsigned long long) PAGE_SHIFT)))
+                        (unsigned long long) PAGE_SHIFT)));     \
+} while (0)
+
 /*
  * Allocate and free page tables.
  */
@@ -32,7 +50,11 @@ static inline void pte_free(struct page *pte)
 }
 
 
-#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
+#define __pte_free_tlb(tlb,pte)                                         \
+do {                                                                    \
+        paravirt_release_pt(page_to_pfn(pte));                          \
+        tlb_remove_page((tlb),(pte));                                   \
+} while (0)
 
 #ifdef CONFIG_X86_PAE
 /*