fedora core 6 1.2949 + vserver 2.2.0
[linux-2.6.git] / include / asm-i386 / mach-xen / asm / pgtable.h
index 388a5ba..7268c13 100644 (file)
@@ -16,6 +16,7 @@
 #include <asm/processor.h>
 #include <asm/fixmap.h>
 #include <linux/threads.h>
+#include <asm/paravirt.h>
 
 #ifndef _I386_BITOPS_H
 #include <asm/bitops.h>
@@ -35,14 +36,14 @@ struct vm_area_struct;
 #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
 extern unsigned long empty_zero_page[1024];
 extern pgd_t *swapper_pg_dir;
-extern kmem_cache_t *pgd_cache;
-extern kmem_cache_t *pmd_cache;
+extern struct kmem_cache *pgd_cache;
+extern struct kmem_cache *pmd_cache;
 extern spinlock_t pgd_lock;
 extern struct page *pgd_list;
 
-void pmd_ctor(void *, kmem_cache_t *, unsigned long);
-void pgd_ctor(void *, kmem_cache_t *, unsigned long);
-void pgd_dtor(void *, kmem_cache_t *, unsigned long);
+void pmd_ctor(void *, struct kmem_cache *, unsigned long);
+void pgd_ctor(void *, struct kmem_cache *, unsigned long);
+void pgd_dtor(void *, struct kmem_cache *, unsigned long);
 void pgtable_cache_init(void);
 void paging_init(void);
 
@@ -249,20 +250,88 @@ static inline pte_t pte_mkhuge(pte_t pte) { (pte).pte_low |= _PAGE_PSE; return p
 # include <asm/pgtable-2level.h>
 #endif
 
-static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
-{
-       if (!pte_dirty(*ptep))
-               return 0;
-       return test_and_clear_bit(_PAGE_BIT_DIRTY, &ptep->pte_low);
-}
+#ifndef CONFIG_PARAVIRT
+/*
+ * Rules for using pte_update - it must be called after any PTE update which
+ * has not been done using the set_pte / clear_pte interfaces.  It is used by
+ * shadow mode hypervisors to resynchronize the shadow page tables.  Kernel PTE
+ * updates should either be sets, clears, or set_pte_atomic for P->P
+ * transitions, which means this hook should only be called for user PTEs.
+ * This hook implies a P->P protection or access change has taken place, which
+ * requires a subsequent TLB flush.  The notification can optionally be delayed
+ * until the TLB flush event by using the pte_update_defer form of the
+ * interface, but care must be taken to assure that the flush happens while
+ * still holding the same page table lock so that the shadow and primary pages
+ * do not become out of sync on SMP.
+ */
+#define pte_update(mm, addr, ptep)             do { } while (0)
+#define pte_update_defer(mm, addr, ptep)       do { } while (0)
+#endif
+
+/*
+ * We only update the dirty/accessed state if we set
+ * the dirty bit by hand in the kernel, since the hardware
+ * will do the accessed bit for us, and we don't want to
+ * race with other CPU's that might be updating the dirty
+ * bit at the same time.
+ */
+#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
+#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
+       do {                                                              \
+               if (__dirty) {                                            \
+                       if ( likely((__vma)->vm_mm == current->mm) ) {    \
+                           BUG_ON(HYPERVISOR_update_va_mapping((__address), (__entry), UVMF_INVLPG|UVMF_MULTI|(unsigned long)((__vma)->vm_mm->cpu_vm_mask.bits))); \
+                           pte_update_defer((__vma)->vm_mm, (__address), (__ptep)); \
+                       } else {                                          \
+                            xen_l1_entry_update((__ptep), (__entry)); \
+                           pte_update_defer((__vma)->vm_mm, (__address), (__ptep)); \
+                           flush_tlb_page((__vma), (__address));         \
+                       }                                                 \
+               }                                                         \
+       } while (0)
+
+/*
+ * We don't actually have these, but we want to advertise them so that
+ * we can encompass the flush here.
+ */
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
+
+#define __HAVE_ARCH_PTEP_CLEAR_DIRTY_FLUSH
+#define ptep_clear_flush_dirty(vma, address, ptep)                     \
+({                                                                     \
+       int __dirty;                                                    \
+       __dirty = pte_dirty(*(ptep));                                   \
+       if (__dirty) {                                                  \
+               clear_bit(_PAGE_BIT_DIRTY, &(ptep)->pte_low);           \
+               pte_update_defer((vma)->vm_mm, (address), (ptep));      \
+               flush_tlb_page(vma, address);                           \
+       }                                                               \
+       __dirty;                                                        \
+})
 
-static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
+#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
+#define ptep_clear_flush_young(vma, address, ptep)                     \
+({                                                                     \
+       int __young;                                                    \
+       __young = pte_young(*(ptep));                                   \
+       if (__young) {                                                  \
+               clear_bit(_PAGE_BIT_ACCESSED, &(ptep)->pte_low);        \
+               pte_update_defer((vma)->vm_mm, (address), (ptep));      \
+               flush_tlb_page(vma, address);                           \
+       }                                                               \
+       __young;                                                        \
+})
+
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
-       if (!pte_young(*ptep))
-               return 0;
-       return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte_low);
+       pte_t pte = raw_ptep_get_and_clear(ptep);
+       pte_update(mm, addr, ptep);
+       return pte;
 }
 
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
 static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
 {
        pte_t pte;
@@ -275,10 +344,12 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
        return pte;
 }
 
+#define __HAVE_ARCH_PTEP_SET_WRPROTECT
 static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
        if (pte_write(*ptep))
                clear_bit(_PAGE_BIT_RW, &ptep->pte_low);
+       pte_update(mm, addr, ptep);
 }
 
 /*
@@ -368,11 +439,11 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_index(address) \
                (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-       ((pte_t *) pmd_page_kernel(*(dir)) +  pte_index(address))
+       ((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
 
 #define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
                ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /*
@@ -395,8 +466,6 @@ extern pte_t *lookup_address(unsigned long address);
  static inline int set_kernel_exec(unsigned long vaddr, int enable) { return 0;}
 #endif
 
-extern void noexec_setup(const char *str);
-
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address) \
        ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE0) + \
@@ -414,29 +483,19 @@ extern void noexec_setup(const char *str);
 #define pte_unmap_nested(pte) do { } while (0)
 #endif
 
+/* Clear a kernel PTE and flush it from the TLB */
+#define kpte_clear_flush(ptep, vaddr)                                  \
+do {                                                                   \
+       pte_clear(&init_mm, vaddr, ptep);                               \
+       __flush_tlb_one(vaddr);                                         \
+} while (0)
+
 /*
  * The i386 doesn't have any external MMU info: the kernel page
  * tables contain all the necessary information.
- *
- * Also, we only update the dirty/accessed state if we set
- * the dirty bit by hand in the kernel, since the hardware
- * will do the accessed bit for us, and we don't want to
- * race with other CPU's that might be updating the dirty
- * bit at the same time.
  */
 #define update_mmu_cache(vma,address,pte) do { } while (0)
 #define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
-#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
-       do {                                                              \
-               if (__dirty) {                                            \
-                       if ( likely((__vma)->vm_mm == current->mm) ) {    \
-                           BUG_ON(HYPERVISOR_update_va_mapping((__address), (__entry), UVMF_INVLPG|UVMF_MULTI|(unsigned long)((__vma)->vm_mm->cpu_vm_mask.bits))); \
-                       } else {                                          \
-                            xen_l1_entry_update((__ptep), (__entry)); \
-                           flush_tlb_page((__vma), (__address));         \
-                       }                                                 \
-               }                                                         \
-       } while (0)
 
 #define __HAVE_ARCH_PTEP_ESTABLISH
 #define ptep_establish(__vma, __address, __ptep, __entry)              \
@@ -490,19 +549,13 @@ int touch_pte_range(struct mm_struct *mm,
                     unsigned long address,
                     unsigned long size);
 
-#define io_remap_pfn_range(vma,from,pfn,size,prot) \
-direct_remap_pfn_range(vma,from,pfn,size,prot,DOMID_IO)
+#define io_remap_pfn_range(vma,vaddr,pfn,size,prot) \
+       direct_remap_pfn_range(vma,vaddr,pfn,size,prot,DOMID_IO)
 
 #define MK_IOSPACE_PFN(space, pfn)     (pfn)
 #define GET_IOSPACE(pfn)               0
 #define GET_PFN(pfn)                   (pfn)
 
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
-#define __HAVE_ARCH_PTEP_SET_WRPROTECT
-#define __HAVE_ARCH_PTE_SAME
 #include <asm-generic/pgtable.h>
 
 #endif /* _I386_PGTABLE_H */