vmalloc中物理地址重映射函数分析

vmalloc中物理地址重映射函数分析

Figure 1:Vmalloc Areas

Figure 2:将非连续3个物理pages映射到连续虚拟地址

注：图片来至《Professional Linux kernel architecture》

函数调用关系：

1) vmap_page_range(unsigned long start, unsigned long end,

pgprot_t prot, struct page **pages)

2) static int vmap_page_range_noflush(unsigned long start, unsigned long end,

pgprot_t prot, struct page **pages)

3) static int vmap_pud_range(pgd_t *pgd, unsigned long addr,

unsigned long end, pgprot_t prot, struct page **pages, int *nr)

4) static int vmap_pmd_range(pud_t *pud, unsigned long addr,

unsigned long end, pgprot_t prot, struct page **pages, int *nr)

//该两个函数实际没做任何事，因为ARM920T只支持2级映射

5) static int vmap_pte_range(pmd_t *pmd, unsigned long addr,

unsigned long end, pgprot_t prot, struct page **pages, int *nr)

//实际填充一级二级页表的函数

6) flush_cache_vmap(start, end);

FIXME:以ARM920T为例，small page(4KiB)映射，MMU只支持2级页表映射，PGD &PTE.

kernel:2.6.35

#ifndef VMALLOC_START

#define VMALLOC_OFFSET (8*1024*1024)

#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1))

#endif

static int vmap_page_range(unsigned long start, unsigned long end,

pgprot_t prot, struct page **pages)

{

int ret;

ret = vmap_page_range_noflush(start, end, prot, pages);

//start,end为开始的虚拟地址，范围在VMALLOC_START ~ VMALLOC_END之间

//VMALLOC_START = high_mem+8MiB

//prot为MMU页表标志位

//pages指向物理页框表的指针（非连续的物理页框）

flush_cache_vmap(start, end);

return ret;

}

static int vmap_page_range_noflush(unsigned long start, unsigned long end,

pgprot_t prot, struct page **pages)

{

pgd_t *pgd;

unsigned long next;

unsigned long addr = start;

int err = 0;

int nr = 0;

BUG_ON(addr >= end);

pgd = pgd_offset_k(addr);

//得到的pgd的地址,pgd的基地址+(addr>>21)，两个连续的一级查找表（2×1MiB）

do {

next = pgd_addr_end(addr, end);

err = vmap_pud_range(pgd, addr, next, prot, pages, &nr);

if (err)

return err;

} while (pgd++, addr = next, addr != end);

return nr;

}

static int vmap_pud_range(pgd_t *pgd, unsigned long addr,

unsigned long end, pgprot_t prot, struct page **pages, int *nr)

{

pud_t *pud;

unsigned long next;

pud = pud_alloc(&init_mm, pgd, addr);

//定义了4level-fixup.h后，pud得到的还是pgd的地址

//#define pud_t pgd_t

if (!pud)

return -ENOMEM;

do {

next = pud_addr_end(addr, end);

if (vmap_pmd_range(pud, addr, next, prot, pages, nr))

return -ENOMEM;

} while (pud++, addr = next, addr != end);

return 0;

}

static int vmap_pmd_range(pud_t *pud, unsigned long addr,

unsigned long end, pgprot_t prot, struct page **pages, int *nr)

{

pmd_t *pmd;

unsigned long next;

pmd = pmd_alloc(&init_mm, pud, addr);

//同pud_alloc，因为只支持2级页表映射，这个函数同样返回pud的地址

if (!pmd)

return -ENOMEM;

do {

next = pmd_addr_end(addr, end);

if (vmap_pte_range(pmd, addr, next, prot, pages, nr))

return -ENOMEM;

} while (pmd++, addr = next, addr != end);

return 0;

}

//vmap映射的最关键函数

static int vmap_pte_range(pmd_t *pmd, unsigned long addr,

unsigned long end, pgprot_t prot, struct page **pages, int *nr)

{

pte_t *pte;

/*

* nr is a running index into the array which helps higher level

* callers keep track of where we're up to.

*/

pte = pte_alloc_kernel(pmd, addr);

//分配二级表地址并返回，并完成一级表的赋值工作，*pmd中填写二级页表的基地址，

//参照create_mapping

if (!pte)

return -ENOMEM;

do {

struct page *page = pages[*nr];

if (WARN_ON(!pte_none(*pte)))

return -EBUSY;

if (WARN_ON(!page))

return -ENOMEM;

set_pte_at(&init_mm, addr, pte, mk_pte(page, prot));

//设置2级页表pte的内容，完成一次Vmap

(*nr)++;

} while (pte++, addr += PAGE_SIZE, addr != end);

return 0;

}

#define pte_alloc_kernel(pmd, address) /

((unlikely(!pmd_present(*(pmd))) && __pte_alloc_kernel(pmd, address))? /

NULL: pte_offset_kernel(pmd, address))

//pmd_present(*(pmd))在未填充时为空，!pmd_present(*(pmd))返回true

int __pte_alloc_kernel(pmd_t *pmd, unsigned long address)

{

pte_t *new = pte_alloc_one_kernel(&init_mm, address);

//利用Buddy allocator分配1page的空间

if (!new)

return -ENOMEM;

smp_wmb(); /* See comment in __pte_alloc */

spin_lock(&init_mm.page_table_lock);

if (!pmd_present(*pmd)) { /* Has another populated it ? */

pmd_populate_kernel(&init_mm, pmd, new);

//给pgd的地址空间赋值，为pte的基地址

new = NULL;

}

spin_unlock(&init_mm.page_table_lock);

if (new)

pte_free_kernel(&init_mm, new);

return 0;

//返回为false

}

Author :woodpecker <Pecker.hu@gmail.com>