【深入理解Linux内核架构】3.3 页表

页表：用于建立用户进程空间的虚拟地址空间和系统物理内存（内存、页帧）之间的关联。

向每个进程提供一致的虚拟地址空间。

将虚拟内存页映射到物理内存，因而支持共享内存的实现。

可以在不增加物理内存的情况下，将页换出到块设备来增加有效的可用内存空间。

内核内存管理总是假定使用四级页表。

3.3.1 数据结构

　　内核源代码假定void *和unsigned long long类型所需的比特位数相同，因此他们可以进行强制转换而不损失信息。即：假定sizeof(void *) == sizeof(unsigned long long)，在Linux支持的所有体系结构上都是正确的。

　　1. 内存地址的分解

　　根据四级页表的结构需求，虚拟地址分为5部分。

　　各个体系结构不仅地址长度不一致，而且地址字拆分的方式也不同。因此内核定义了宏，用于将地址分解为各个分量。

　　BITS_PER_LONG定义用于unsigned long变量的比特位数目，因而也适用于指向虚拟地址空间的通用指针。

　　关于上述设计的宏例如PAGE_SHIFT的定义，是在文件page.h中定义的。在linux下，page.h的定义有2个地方，一个是linux-3.08/include/asm-generic/page.h，一个是在架构相关的目录，如mips是在linux-3.08/arch/mips/include/asm/page.h。一般而言，如果架构目录定义那么肯定会使用架构目录下的定义。

　　所以我们看看 linux-3.08/arch/mips/include/asm/page.h文件：

/*

 * This file is subject to the terms and conditions of the GNU General Public

 * License.  See the file "COPYING" in the main directory of this archive

 * for more details.

 *

 * Copyright (C) 1994 - 1999, 2000, 03 Ralf Baechle

 * Copyright (C) 1999, 2000 Silicon Graphics, Inc.

 */

#ifndef _ASM_PAGE_H

#define _ASM_PAGE_H

#include <spaces.h>

#include <linux/const.h>

/*

 * PAGE_SHIFT determines the page size

 */

#ifdef CONFIG_PAGE_SIZE_4KB

#define PAGE_SHIFT    12

#endif

#ifdef CONFIG_PAGE_SIZE_8KB

#define PAGE_SHIFT    13

#endif

#ifdef CONFIG_PAGE_SIZE_16KB

#define PAGE_SHIFT    14

#endif

#ifdef CONFIG_PAGE_SIZE_32KB

#define PAGE_SHIFT    15

#endif

#ifdef CONFIG_PAGE_SIZE_64KB

#define PAGE_SHIFT    16

#endif

#define PAGE_SIZE    (_AC(1,UL) << PAGE_SHIFT)

#define PAGE_MASK       (~((1 << PAGE_SHIFT) - 1))

#ifdef CONFIG_HUGETLB_PAGE

#define HPAGE_SHIFT    (PAGE_SHIFT + PAGE_SHIFT - 3)

#define HPAGE_SIZE    (_AC(1,UL) << HPAGE_SHIFT)

#define HPAGE_MASK    (~(HPAGE_SIZE - 1))

#define HUGETLB_PAGE_ORDER    (HPAGE_SHIFT - PAGE_SHIFT)

#endif /* CONFIG_HUGETLB_PAGE */

#ifndef __ASSEMBLY__

#include <linux/pfn.h>

#include <asm/io.h>

extern void build_clear_page(void);

extern void build_copy_page(void);

/*

 * It's normally defined only for FLATMEM config but it's

 * used in our early mem init code for all memory models.

 * So always define it.

 */

#define ARCH_PFN_OFFSET        PFN_UP(PHYS_OFFSET)

extern void clear_page(void * page);

extern void copy_page(void * to, void * from);

extern unsigned long shm_align_mask;

static inline unsigned long pages_do_alias(unsigned long addr1,

    unsigned long addr2)

{

    return (addr1 ^ addr2) & shm_align_mask;

}

struct page;

static inline void clear_user_page(void *addr, unsigned long vaddr,

    struct page *page)

{

    extern void (*flush_data_cache_page)(unsigned long addr);

    clear_page(addr);

    if (cpu_has_vtag_dcache || (cpu_has_dc_aliases &&

         pages_do_alias((unsigned long) addr, vaddr & PAGE_MASK)))

        flush_data_cache_page((unsigned long)addr);

}

extern void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,

    struct page *to);

struct vm_area_struct;

extern void copy_user_highpage(struct page *to, struct page *from,

    unsigned long vaddr, struct vm_area_struct *vma);

#define __HAVE_ARCH_COPY_USER_HIGHPAGE

/*

 * These are used to make use of C type-checking..

 */

#ifdef CONFIG_64BIT_PHYS_ADDR

  #ifdef CONFIG_CPU_MIPS32

    typedef struct { unsigned long pte_low, pte_high; } pte_t;

    #define pte_val(x)    ((x).pte_low | ((unsigned long long)(x).pte_high << 32))

    #define __pte(x)      ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; })

  #else

     typedef struct { unsigned long long pte; } pte_t;

     #define pte_val(x)    ((x).pte)

     #define __pte(x)    ((pte_t) { (x) } )

  #endif

#else

typedef struct { unsigned long pte; } pte_t;

#define pte_val(x)    ((x).pte)

#define __pte(x)    ((pte_t) { (x) } )

#endif

typedef struct page *pgtable_t;

/*

 * Right now we don't support 4-level pagetables, so all pud-related

 * definitions come from <asm-generic/pgtable-nopud.h>.

 */

/*

 * Finall the top of the hierarchy, the pgd

 */

typedef struct { unsigned long pgd; } pgd_t;

#define pgd_val(x)    ((x).pgd)

#define __pgd(x)    ((pgd_t) { (x) } )

/*

 * Manipulate page protection bits

 */

typedef struct { unsigned long pgprot; } pgprot_t;

#define pgprot_val(x)    ((x).pgprot)

#define __pgprot(x)    ((pgprot_t) { (x) } )

/*

 * On R4000-style MMUs where a TLB entry is mapping a adjacent even / odd

 * pair of pages we only have a single global bit per pair of pages.  When

 * writing to the TLB make sure we always have the bit set for both pages

 * or none.  This macro is used to access the `buddy' of the pte we're just

 * working on.

 */

#define ptep_buddy(x)    ((pte_t *)((unsigned long)(x) ^ sizeof(pte_t)))

#endif /* !__ASSEMBLY__ */

/*

 * __pa()/__va() should be used only during mem init.

 */

#ifdef CONFIG_64BIT

#define __pa(x)                                \

({                                    \

    unsigned long __x = (unsigned long)(x);                \

    __x < CKSEG0 ? XPHYSADDR(__x) : CPHYSADDR(__x);            \

})

#else

#define __pa(x)                                \

    ((unsigned long)(x) - PAGE_OFFSET + PHYS_OFFSET)

#endif

#define __va(x)        ((void *)((unsigned long)(x) + PAGE_OFFSET - PHYS_OFFSET))

/*

 * RELOC_HIDE was originally added by 6007b903dfe5f1d13e0c711ac2894bdd4a61b1ad

 * (lmo) rsp. 8431fd094d625b94d364fe393076ccef88e6ce18 (kernel.org).  The

 * discussion can be found in lkml posting

 * <a2ebde260608230500o3407b108hc03debb9da6e62c@mail.gmail.com> which is

 * archived at http://lists.linuxcoding.com/kernel/2006-q3/msg17360.html

 *

 * It is unclear if the misscompilations mentioned in

 * http://lkml.org/lkml/2010/8/8/138 also affect MIPS so we keep this one

 * until GCC 3.x has been retired before we can apply

 * https://patchwork.linux-mips.org/patch/1541/

 */

#define __pa_symbol(x)    __pa(RELOC_HIDE((unsigned long)(x), 0))

#define pfn_to_kaddr(pfn)    __va((pfn) << PAGE_SHIFT)

#ifdef CONFIG_FLATMEM

#define pfn_valid(pfn)                            \

({                                    \

    unsigned long __pfn = (pfn);                    \

    /* avoid <linux/bootmem.h> include hell */            \

    extern unsigned long min_low_pfn;                \

                                    \

    __pfn >= min_low_pfn && __pfn < max_mapnr;            \

})

#elif defined(CONFIG_SPARSEMEM)

/* pfn_valid is defined in linux/mmzone.h */

#elif defined(CONFIG_NEED_MULTIPLE_NODES)

#define pfn_valid(pfn)                            \

({                                    \

    unsigned long __pfn = (pfn);                    \

    int __n = pfn_to_nid(__pfn);                    \

    ((__n >= ) ? (__pfn < NODE_DATA(__n)->node_start_pfn +        \

                           NODE_DATA(__n)->node_spanned_pages)    \

                : );                        \

})

#endif

#define virt_to_page(kaddr)    pfn_to_page(PFN_DOWN(virt_to_phys(kaddr)))

#define virt_addr_valid(kaddr)    pfn_valid(PFN_DOWN(virt_to_phys(kaddr)))

/*

#define VM_DATA_DEFAULT_FLAGS    (VM_READ | VM_WRITE | VM_EXEC | \

                 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)

*/

#define VM_DATA_DEFAULT_FLAGS    (VM_READ | VM_WRITE | \

                 VM_MAYREAD | VM_MAYWRITE)

#define UNCAC_ADDR(addr)    ((addr) - PAGE_OFFSET + UNCAC_BASE +     \

                                PHYS_OFFSET)

#define CAC_ADDR(addr)        ((addr) - UNCAC_BASE + PAGE_OFFSET -    \

                                PHYS_OFFSET)

#include <asm-generic/memory_model.h>

#include <asm-generic/getorder.h>

#endif /* _ASM_PAGE_H */

　　阅读上述代码可以得到以下事实：

　　　　　　PAGE_SHIFT：最后一级页表项所需比特位的总是。对于32位系统，PAGE_SHIFT==12.

　　　　　　PAGE_SIZE：一页的大小。对32位系统，PAGE_SIZE == 4096.

xxxx  xxxx xxxx  xxxx  xxxx  xxxx xxxx  xxxx

          ( << )        【PAGE_SIZE】

          (( << ) -)

          (~(( << ) -))【PAGE_MASK】

　　对于mips32架构，关于PGDIR_SHIFT的定义：linux-3.08/arch/mips/include/asm/pgtable-32.h

/*

 * This file is subject to the terms and conditions of the GNU General Public

 * License.  See the file "COPYING" in the main directory of this archive

 * for more details.

 *

 * Copyright (C) 1994, 95, 96, 97, 98, 99, 2000, 2003 Ralf Baechle

 * Copyright (C) 1999, 2000, 2001 Silicon Graphics, Inc.

 */

#ifndef _ASM_PGTABLE_32_H

#define _ASM_PGTABLE_32_H

#include <asm/addrspace.h>

#include <asm/page.h>

#include <linux/linkage.h>

#include <asm/cachectl.h>

#include <asm/fixmap.h>

#include <asm-generic/pgtable-nopmd.h>

/*

 * - add_wired_entry() add a fixed TLB entry, and move wired register

 */

extern void add_wired_entry(unsigned long entrylo0, unsigned long entrylo1,

                   unsigned long entryhi, unsigned long pagemask);

/*

 * - add_temporary_entry() add a temporary TLB entry. We use TLB entries

 *    starting at the top and working down. This is for populating the

 *    TLB before trap_init() puts the TLB miss handler in place. It

 *    should be used only for entries matching the actual page tables,

 *    to prevent inconsistencies.

 */

extern int add_temporary_entry(unsigned long entrylo0, unsigned long entrylo1,

                   unsigned long entryhi, unsigned long pagemask);

/* Basically we have the same two-level (which is the logical three level

 * Linux page table layout folded) page tables as the i386.  Some day

 * when we have proper page coloring support we can have a 1% quicker

 * tlb refill handling mechanism, but for now it is a bit slower but

 * works even with the cache aliasing problem the R4k and above have.

 */

/* PGDIR_SHIFT determines what a third-level page table entry can map */

#define PGDIR_SHIFT    (2 * PAGE_SHIFT + PTE_ORDER - PTE_T_LOG2)

#define PGDIR_SIZE    (1UL << PGDIR_SHIFT)

#define PGDIR_MASK    (~(PGDIR_SIZE-1))

/*

 * Entries per page directory level: we use two-level, so

 * we don't really have any PUD/PMD directory physically.

 */

#define __PGD_ORDER    (32 - 3 * PAGE_SHIFT + PGD_T_LOG2 + PTE_T_LOG2)

#define PGD_ORDER    (__PGD_ORDER >= 0 ? __PGD_ORDER : 0)

#define PUD_ORDER    aieeee_attempt_to_allocate_pud

#define PMD_ORDER    1

#define PTE_ORDER    0

#define PTRS_PER_PGD    (USER_PTRS_PER_PGD * 2)

#define PTRS_PER_PTE    ((PAGE_SIZE << PTE_ORDER) / sizeof(pte_t))

#define USER_PTRS_PER_PGD    (0x80000000UL/PGDIR_SIZE)

#define FIRST_USER_ADDRESS    0

#define VMALLOC_START     MAP_BASE

#define PKMAP_BASE        (0xfe000000UL)

#ifdef CONFIG_HIGHMEM

# define VMALLOC_END    (PKMAP_BASE-*PAGE_SIZE)

#else

# define VMALLOC_END    (FIXADDR_START-*PAGE_SIZE)

#endif

#ifdef CONFIG_64BIT_PHYS_ADDR

#define pte_ERROR(e) \

    printk("%s:%d: bad pte %016Lx.\n", __FILE__, __LINE__, pte_val(e))

#else

#define pte_ERROR(e) \

    printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))

#endif

#define pgd_ERROR(e) \

    printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))

extern void load_pgd(unsigned long pg_dir);

extern pte_t invalid_pte_table[PAGE_SIZE/sizeof(pte_t)];

/*

 * Empty pgd/pmd entries point to the invalid_pte_table.

 */

static inline int pmd_none(pmd_t pmd)

{

    return pmd_val(pmd) == (unsigned long) invalid_pte_table;

}

#define pmd_bad(pmd)        (pmd_val(pmd) & ~PAGE_MASK)

static inline int pmd_present(pmd_t pmd)

{

    return pmd_val(pmd) != (unsigned long) invalid_pte_table;

}

static inline void pmd_clear(pmd_t *pmdp)

{

    pmd_val(*pmdp) = ((unsigned long) invalid_pte_table);

}

#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32)

#define pte_page(x)        pfn_to_page(pte_pfn(x))

#define pte_pfn(x)        ((unsigned long)((x).pte_high >> 6))

static inline pte_t

pfn_pte(unsigned long pfn, pgprot_t prot)

{

    pte_t pte;

    pte.pte_high = (pfn << ) | (pgprot_val(prot) & 0x3f);

    pte.pte_low = pgprot_val(prot);

    return pte;

}

#else

#define pte_page(x)        pfn_to_page(pte_pfn(x))

#ifdef CONFIG_CPU_VR41XX

#define pte_pfn(x)        ((unsigned long)((x).pte >> (PAGE_SHIFT + 2)))

#define pfn_pte(pfn, prot)    __pte(((pfn) << (PAGE_SHIFT + 2)) | pgprot_val(prot))

#else

#define pte_pfn(x)        ((unsigned long)((x).pte >> _PFN_SHIFT))

#define pfn_pte(pfn, prot)    __pte(((unsigned long long)(pfn) << _PFN_SHIFT) | pgprot_val(prot))

#endif

#endif /* defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32) */

#define __pgd_offset(address)    pgd_index(address)

#define __pud_offset(address)    (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1))

#define __pmd_offset(address)    (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))

/* to find an entry in a kernel page-table-directory */

#define pgd_offset_k(address) pgd_offset(&init_mm, address)

#define pgd_index(address)    (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))

/* to find an entry in a page-table-directory */

#define pgd_offset(mm, addr)    ((mm)->pgd + pgd_index(addr))

/* Find an entry in the third-level page table.. */

#define __pte_offset(address)                        \

    (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - ))

#define pte_offset(dir, address)                    \

    ((pte_t *) pmd_page_vaddr(*(dir)) + __pte_offset(address))

#define pte_offset_kernel(dir, address)                    \

    ((pte_t *) pmd_page_vaddr(*(dir)) + __pte_offset(address))

#define pte_offset_map(dir, address)                                    \

    ((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))

#define pte_unmap(pte) ((void)(pte))

#if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)

/* Swap entries must have VALID bit cleared. */

#define __swp_type(x)        (((x).val >> 10) & 0x1f)

#define __swp_offset(x)        ((x).val >> 15)

#define __swp_entry(type,offset)    \

    ((swp_entry_t) { ((type) << ) | ((offset) << ) })

/*

 * Bits 0, 4, 8, and 9 are taken, split up 28 bits of offset into this range:

 */

#define PTE_FILE_MAX_BITS    28

#define pte_to_pgoff(_pte)    ((((_pte).pte >> 1 ) & 0x07) | \

                 (((_pte).pte >>  ) & 0x38) | \

                 (((_pte).pte >> ) <<   ))

#define pgoff_to_pte(off)    ((pte_t) { (((off) & 0x07) << 1 ) | \

                       (((off) & 0x38) <<  ) | \

                       (((off) >>   ) << ) | \

                       _PAGE_FILE })

#else

/* Swap entries must have VALID and GLOBAL bits cleared. */

#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32)

#define __swp_type(x)        (((x).val >> 2) & 0x1f)

#define __swp_offset(x)      ((x).val >> 7)

#define __swp_entry(type,offset)    \

        ((swp_entry_t)  { ((type) << ) | ((offset) << ) })

#else

#define __swp_type(x)        (((x).val >> 8) & 0x1f)

#define __swp_offset(x)      ((x).val >> 13)

#define __swp_entry(type,offset)    \

        ((swp_entry_t)  { ((type) << ) | ((offset) << ) })

#endif /* defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32) */

#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32)

/*

 * Bits 0 and 1 of pte_high are taken, use the rest for the page offset...

 */

#define PTE_FILE_MAX_BITS    30

#define pte_to_pgoff(_pte)    ((_pte).pte_high >> 2)

#define pgoff_to_pte(off)     ((pte_t) { _PAGE_FILE, (off) << 2 })

#else

/*

 * Bits 0, 4, 6, and 7 are taken, split up 28 bits of offset into this range:

 */

#define PTE_FILE_MAX_BITS    28

#define pte_to_pgoff(_pte)    ((((_pte).pte >> 1) & 0x7) | \

                 (((_pte).pte >> ) & 0x8) | \

                 (((_pte).pte >> ) <<  ))

#define pgoff_to_pte(off)    ((pte_t) { (((off) & 0x7) << 1) | \

                       (((off) & 0x8) << ) | \

                       (((off) >>  ) << ) | \

                       _PAGE_FILE })

#endif

#endif

#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32)

#define __pte_to_swp_entry(pte) ((swp_entry_t) { (pte).pte_high })

#define __swp_entry_to_pte(x)    ((pte_t) { 0, (x).val })

#else

#define __pte_to_swp_entry(pte)    ((swp_entry_t) { pte_val(pte) })

#define __swp_entry_to_pte(x)    ((pte_t) { (x).val })

#endif

#endif /* _ASM_PGTABLE_32_H */

　　我只关心mips32架构的相关配置，所以，此时，对于四级页表的Linux，如何为mips32设置成3级页表？

　　Linux提供了通用的没有PUD和PMD的相关的配置。配置文件：linux-3.08/include/asm-generic/pgtable-nopmd.h和pgtable-nopud.h:

/* linux-3.08/include/asm-generic/pgtable-nopmd.h */

#ifndef _PGTABLE_NOPMD_H

#define _PGTABLE_NOPMD_H

#ifndef __ASSEMBLY__

#include <asm-generic/pgtable-nopud.h>

struct mm_struct;

#define __PAGETABLE_PMD_FOLDED

/*

 * Having the pmd type consist of a pud gets the size right, and allows

 * us to conceptually access the pud entry that this pmd is folded into

 * without casting.

 */

typedef struct { pud_t pud; } pmd_t;

#define PMD_SHIFT    PUD_SHIFT

#define PTRS_PER_PMD    1

#define PMD_SIZE      (1UL << PMD_SHIFT)

#define PMD_MASK      (~(PMD_SIZE-1))

/*

 * The "pud_xxx()" functions here are trivial for a folded two-level

 * setup: the pmd is never bad, and a pmd always exists (as it's folded

 * into the pud entry)

 */

static inline int pud_none(pud_t pud)        { return ; }

static inline int pud_bad(pud_t pud)        { return ; }

static inline int pud_present(pud_t pud)    { return ; }

static inline void pud_clear(pud_t *pud)    { }

#define pmd_ERROR(pmd)                (pud_ERROR((pmd).pud))

#define pud_populate(mm, pmd, pte)        do { } while (0)

/*

 * (pmds are folded into puds so this doesn't get actually called,

 * but the define is needed for a generic inline function.)

 */

#define set_pud(pudptr, pudval)            set_pmd((pmd_t *)(pudptr), (pmd_t) { pudval })

static inline pmd_t * pmd_offset(pud_t * pud, unsigned long address)

{

    return (pmd_t *)pud;

}

#define pmd_val(x)                (pud_val((x).pud))

#define __pmd(x)                ((pmd_t) { __pud(x) } )

#define pud_page(pud)                (pmd_page((pmd_t){ pud }))

#define pud_page_vaddr(pud)            (pmd_page_vaddr((pmd_t){ pud }))

/*

 * allocating and freeing a pmd is trivial: the 1-entry pmd is

 * inside the pud, so has no extra memory associated with it.

 */

#define pmd_alloc_one(mm, address)        NULL

static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd)

{

}

#define __pmd_free_tlb(tlb, x, a)        do { } while (0)

#undef  pmd_addr_end

#define pmd_addr_end(addr, end)            (end)

#endif /* __ASSEMBLY__ */

#endif /* _PGTABLE_NOPMD_H */

/* linux-3.08/include/asm-generic/pgtable-nopud.h */

#ifndef _PGTABLE_NOPUD_H

#define _PGTABLE_NOPUD_H

#ifndef __ASSEMBLY__

#define __PAGETABLE_PUD_FOLDED

/*

 * Having the pud type consist of a pgd gets the size right, and allows

 * us to conceptually access the pgd entry that this pud is folded into

 * without casting.

 */

typedef struct { pgd_t pgd; } pud_t;

#define PUD_SHIFT    PGDIR_SHIFT

#define PTRS_PER_PUD    1

#define PUD_SIZE      (1UL << PUD_SHIFT)

#define PUD_MASK      (~(PUD_SIZE-1))

/*

 * The "pgd_xxx()" functions here are trivial for a folded two-level

 * setup: the pud is never bad, and a pud always exists (as it's folded

 * into the pgd entry)

 */

static inline int pgd_none(pgd_t pgd)        { return ; }

static inline int pgd_bad(pgd_t pgd)        { return ; }

static inline int pgd_present(pgd_t pgd)    { return ; }

static inline void pgd_clear(pgd_t *pgd)    { }

#define pud_ERROR(pud)                (pgd_ERROR((pud).pgd))

#define pgd_populate(mm, pgd, pud)        do { } while (0)

/*

 * (puds are folded into pgds so this doesn't get actually called,

 * but the define is needed for a generic inline function.)

 */

#define set_pgd(pgdptr, pgdval)            set_pud((pud_t *)(pgdptr), (pud_t) { pgdval })

static inline pud_t * pud_offset(pgd_t * pgd, unsigned long address)

{

    return (pud_t *)pgd;

}

#define pud_val(x)                (pgd_val((x).pgd))

#define __pud(x)                ((pud_t) { __pgd(x) } )

#define pgd_page(pgd)                (pud_page((pud_t){ pgd }))

#define pgd_page_vaddr(pgd)            (pud_page_vaddr((pud_t){ pgd }))

/*

 * allocating and freeing a pud is trivial: the 1-entry pud is

 * inside the pgd, so has no extra memory associated with it.

 */

#define pud_alloc_one(mm, address)        NULL

#define pud_free(mm, x)                do { } while (0)

#define __pud_free_tlb(tlb, x, a)        do { } while (0)

#undef  pud_addr_end

#define pud_addr_end(addr, end)            (end)

#endif /* __ASSEMBLY__ */

#endif /* _PGTABLE_NOPUD_H */

　　尽管：PMD_SHIFT和PUD_SHIFT都定义为PGDIR_SHIFT。但是关键字：PDRS_PER_PUD和PTRS_PER_PMD都定义为1。那么意义是什么呢？

　　　　 PDRS_PER_PUD：指定了二级页表（PUD）所能存储的指针数目。

　　　　 PDRS_PED_PMD：指定了三级页表（PMD）所能存储的指针数目。

　　　　设置为1，Linux内核还是以为是四级页表，但是实际上只有二级页表。

在PAGE_SHIFT代码中，我们看到了PAGE_MASK。类似存在：PUD_MASK、PMD_MASK、PGDIR_MASK。

那么这些MASK的作用是：从给定地址中提取各个分量。【用给定地址与对应的MASK位与即可获得各个分量】

2. 页表的格式

　　pgd_t:全局页目录项。

　　pud_t:上层页目录项。

　　pmd_t:中间页目录项。

　　pte_t：直接页表项。

typedef struct { unsigned long pgd; } pgd_t;

typedef struct { pgd_t pgd; } pud_t;

typedef struct { pgd_t pud; } pmd_t;

typedef struct { unsigned long pte; } pte_t;

PAGE_ALIGN:将输入的地址对其到下一页的起始处。如页大小是4096，该宏总返回其倍数。PAGE_ALIGN(6000) = 8192.

3. 特定于PTE的信息

　　最后一级页表中的项不仅包含了只想页的内存位置的指针，还在上述的多余的比特位包含了与页有关的附加信息。这些信息特定于CPU，提供了页的访问控制信息。

　　详细内容不再细述。用到时我们再回头来看。

巴特西

【深入理解Linux内核架构】3.3 页表

最新文章

热门文章