2005-04-16 16:20:36 -06:00
|
|
|
#ifndef _X86_64_PGTABLE_H
|
|
|
|
#define _X86_64_PGTABLE_H
|
|
|
|
|
2007-05-08 01:31:11 -06:00
|
|
|
#include <linux/const.h>
|
2007-05-02 11:27:06 -06:00
|
|
|
#ifndef __ASSEMBLY__
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* This file contains the functions and defines necessary to modify and use
|
|
|
|
* the x86-64 page table tree.
|
|
|
|
*/
|
|
|
|
#include <asm/processor.h>
|
2007-10-19 00:40:25 -06:00
|
|
|
#include <linux/bitops.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/threads.h>
|
|
|
|
#include <asm/pda.h>
|
|
|
|
|
|
|
|
extern pud_t level3_kernel_pgt[512];
|
|
|
|
extern pud_t level3_ident_pgt[512];
|
|
|
|
extern pmd_t level2_kernel_pgt[512];
|
|
|
|
extern pgd_t init_level4_pgt[];
|
|
|
|
|
Revert "[PATCH] x86: __pa and __pa_symbol address space separation"
This was broken. It adds complexity, for no good reason. Rather than
separate __pa() and __pa_symbol(), we should deprecate __pa_symbol(),
and preferably __pa() too - and just use "virt_to_phys()" instead, which
is more readable and has nicer semantics.
However, right now, just undo the separation, and make __pa_symbol() be
the exact same as __pa(). That fixes the bugs this patch introduced,
and we can do the fairly obvious cleanups later.
Do the new __phys_addr() function (which is now the actual workhorse for
the unified __pa()/__pa_symbol()) as a real external function, that way
all the potential issues with compile/link-time optimizations of
constant symbol addresses go away, and we can also, if we choose to, add
more sanity-checking of the argument.
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-07 09:44:24 -06:00
|
|
|
#define swapper_pg_dir init_level4_pgt
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
extern void paging_init(void);
|
|
|
|
extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* ZERO_PAGE is a global shared page that is always zero: used
|
|
|
|
* for zero-mapped memory areas etc..
|
|
|
|
*/
|
|
|
|
extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
|
Revert "[PATCH] x86: __pa and __pa_symbol address space separation"
This was broken. It adds complexity, for no good reason. Rather than
separate __pa() and __pa_symbol(), we should deprecate __pa_symbol(),
and preferably __pa() too - and just use "virt_to_phys()" instead, which
is more readable and has nicer semantics.
However, right now, just undo the separation, and make __pa_symbol() be
the exact same as __pa(). That fixes the bugs this patch introduced,
and we can do the fairly obvious cleanups later.
Do the new __phys_addr() function (which is now the actual workhorse for
the unified __pa()/__pa_symbol()) as a real external function, that way
all the potential issues with compile/link-time optimizations of
constant symbol addresses go away, and we can also, if we choose to, add
more sanity-checking of the argument.
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-07 09:44:24 -06:00
|
|
|
#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2007-05-02 11:27:06 -06:00
|
|
|
#endif /* !__ASSEMBLY__ */
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* PGDIR_SHIFT determines what a top-level page table entry can map
|
|
|
|
*/
|
|
|
|
#define PGDIR_SHIFT 39
|
|
|
|
#define PTRS_PER_PGD 512
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 3rd level page
|
|
|
|
*/
|
|
|
|
#define PUD_SHIFT 30
|
|
|
|
#define PTRS_PER_PUD 512
|
|
|
|
|
|
|
|
/*
|
|
|
|
* PMD_SHIFT determines the size of the area a middle-level
|
|
|
|
* page table can map
|
|
|
|
*/
|
|
|
|
#define PMD_SHIFT 21
|
|
|
|
#define PTRS_PER_PMD 512
|
|
|
|
|
|
|
|
/*
|
|
|
|
* entries per page directory level
|
|
|
|
*/
|
|
|
|
#define PTRS_PER_PTE 512
|
|
|
|
|
2007-05-02 11:27:06 -06:00
|
|
|
#ifndef __ASSEMBLY__
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
#define pte_ERROR(e) \
|
|
|
|
printk("%s:%d: bad pte %p(%016lx).\n", __FILE__, __LINE__, &(e), pte_val(e))
|
|
|
|
#define pmd_ERROR(e) \
|
|
|
|
printk("%s:%d: bad pmd %p(%016lx).\n", __FILE__, __LINE__, &(e), pmd_val(e))
|
|
|
|
#define pud_ERROR(e) \
|
|
|
|
printk("%s:%d: bad pud %p(%016lx).\n", __FILE__, __LINE__, &(e), pud_val(e))
|
|
|
|
#define pgd_ERROR(e) \
|
|
|
|
printk("%s:%d: bad pgd %p(%016lx).\n", __FILE__, __LINE__, &(e), pgd_val(e))
|
|
|
|
|
|
|
|
#define pgd_none(x) (!pgd_val(x))
|
|
|
|
#define pud_none(x) (!pud_val(x))
|
|
|
|
|
|
|
|
static inline void set_pte(pte_t *dst, pte_t val)
|
|
|
|
{
|
|
|
|
pte_val(*dst) = pte_val(val);
|
|
|
|
}
|
|
|
|
#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
|
|
|
|
|
|
|
|
static inline void set_pmd(pmd_t *dst, pmd_t val)
|
|
|
|
{
|
2008-01-30 05:32:42 -07:00
|
|
|
*dst = val;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_pud(pud_t *dst, pud_t val)
|
|
|
|
{
|
2008-01-30 05:32:42 -07:00
|
|
|
*dst = val;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2005-09-12 10:49:24 -06:00
|
|
|
static inline void pud_clear (pud_t *pud)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
set_pud(pud, __pud(0));
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_pgd(pgd_t *dst, pgd_t val)
|
|
|
|
{
|
2008-01-30 05:32:42 -07:00
|
|
|
*dst = val;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2005-09-12 10:49:24 -06:00
|
|
|
static inline void pgd_clear (pgd_t * pgd)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
set_pgd(pgd, __pgd(0));
|
|
|
|
}
|
|
|
|
|
|
|
|
#define ptep_get_and_clear(mm,addr,xp) __pte(xchg(&(xp)->pte, 0))
|
2005-09-03 16:55:06 -06:00
|
|
|
|
2005-11-07 01:59:43 -07:00
|
|
|
struct mm_struct;
|
|
|
|
|
2005-09-03 16:55:06 -06:00
|
|
|
static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
|
|
|
|
{
|
|
|
|
pte_t pte;
|
|
|
|
if (full) {
|
|
|
|
pte = *ptep;
|
|
|
|
*ptep = __pte(0);
|
|
|
|
} else {
|
|
|
|
pte = ptep_get_and_clear(mm, addr, ptep);
|
|
|
|
}
|
|
|
|
return pte;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
#define pte_same(a, b) ((a).pte == (b).pte)
|
|
|
|
|
2006-01-06 01:12:03 -07:00
|
|
|
#define pte_pgprot(a) (__pgprot((a).pte & ~PHYSICAL_PAGE_MASK))
|
|
|
|
|
2007-05-02 11:27:06 -06:00
|
|
|
#endif /* !__ASSEMBLY__ */
|
|
|
|
|
|
|
|
#define PMD_SIZE (_AC(1,UL) << PMD_SHIFT)
|
2005-04-16 16:20:36 -06:00
|
|
|
#define PMD_MASK (~(PMD_SIZE-1))
|
2007-05-02 11:27:06 -06:00
|
|
|
#define PUD_SIZE (_AC(1,UL) << PUD_SHIFT)
|
2005-04-16 16:20:36 -06:00
|
|
|
#define PUD_MASK (~(PUD_SIZE-1))
|
2007-05-02 11:27:06 -06:00
|
|
|
#define PGDIR_SIZE (_AC(1,UL) << PGDIR_SHIFT)
|
2005-04-16 16:20:36 -06:00
|
|
|
#define PGDIR_MASK (~(PGDIR_SIZE-1))
|
|
|
|
|
|
|
|
|
2007-05-08 01:31:14 -06:00
|
|
|
#define MAXMEM _AC(0x3fffffffffff, UL)
|
|
|
|
#define VMALLOC_START _AC(0xffffc20000000000, UL)
|
|
|
|
#define VMALLOC_END _AC(0xffffe1ffffffffff, UL)
|
2007-10-16 02:24:15 -06:00
|
|
|
#define VMEMMAP_START _AC(0xffffe20000000000, UL)
|
2007-05-08 01:31:14 -06:00
|
|
|
#define MODULES_VADDR _AC(0xffffffff88000000, UL)
|
|
|
|
#define MODULES_END _AC(0xfffffffffff00000, UL)
|
2005-04-16 16:20:36 -06:00
|
|
|
#define MODULES_LEN (MODULES_END - MODULES_VADDR)
|
|
|
|
|
2007-05-02 11:27:06 -06:00
|
|
|
#ifndef __ASSEMBLY__
|
|
|
|
|
2006-12-06 18:14:09 -07:00
|
|
|
static inline unsigned long pgd_bad(pgd_t pgd)
|
|
|
|
{
|
|
|
|
return pgd_val(pgd) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER);
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
static inline unsigned long pud_bad(pud_t pud)
|
|
|
|
{
|
2006-12-06 18:14:09 -07:00
|
|
|
return pud_val(pud) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline unsigned long pmd_bad(pmd_t pmd)
|
|
|
|
{
|
|
|
|
return pmd_val(pmd) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
#define pte_none(x) (!pte_val(x))
|
|
|
|
#define pte_present(x) (pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
|
|
|
|
#define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
|
|
|
|
|
2008-01-30 05:31:02 -07:00
|
|
|
#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT)) /* FIXME: is this right? */
|
2005-04-16 16:20:36 -06:00
|
|
|
#define pte_page(x) pfn_to_page(pte_pfn(x))
|
2005-11-05 09:25:53 -07:00
|
|
|
#define pte_pfn(x) ((pte_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
struct vm_area_struct;
|
|
|
|
|
|
|
|
static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
|
|
|
|
{
|
|
|
|
if (!pte_young(*ptep))
|
|
|
|
return 0;
|
2006-03-24 04:15:11 -07:00
|
|
|
return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
|
|
|
|
{
|
2006-03-24 04:15:11 -07:00
|
|
|
clear_bit(_PAGE_BIT_RW, &ptep->pte);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Macro to mark a page protection value as "uncacheable".
|
|
|
|
*/
|
|
|
|
#define pgprot_noncached(prot) (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Conversion functions: convert a page and protection to a page entry,
|
|
|
|
* and a page entry and page directory to the page they refer to.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Level 4 access.
|
|
|
|
*/
|
2006-09-26 00:31:48 -06:00
|
|
|
#define pgd_page_vaddr(pgd) ((unsigned long) __va((unsigned long)pgd_val(pgd) & PTE_MASK))
|
|
|
|
#define pgd_page(pgd) (pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT))
|
2005-04-16 16:20:36 -06:00
|
|
|
#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
|
|
|
|
#define pgd_offset(mm, addr) ((mm)->pgd + pgd_index(addr))
|
|
|
|
#define pgd_offset_k(address) (init_level4_pgt + pgd_index(address))
|
|
|
|
#define pgd_present(pgd) (pgd_val(pgd) & _PAGE_PRESENT)
|
|
|
|
#define mk_kernel_pgd(address) ((pgd_t){ (address) | _KERNPG_TABLE })
|
|
|
|
|
|
|
|
/* PUD - Level3 access */
|
|
|
|
/* to find an entry in a page-table-directory. */
|
2006-09-26 00:31:48 -06:00
|
|
|
#define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
|
|
|
|
#define pud_page(pud) (pfn_to_page(pud_val(pud) >> PAGE_SHIFT))
|
2005-04-16 16:20:36 -06:00
|
|
|
#define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1))
|
2006-09-26 00:31:48 -06:00
|
|
|
#define pud_offset(pgd, address) ((pud_t *) pgd_page_vaddr(*(pgd)) + pud_index(address))
|
2005-04-16 16:20:36 -06:00
|
|
|
#define pud_present(pud) (pud_val(pud) & _PAGE_PRESENT)
|
|
|
|
|
|
|
|
/* PMD - Level 2 access */
|
2006-09-26 00:31:48 -06:00
|
|
|
#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
|
2005-04-16 16:20:36 -06:00
|
|
|
#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
|
|
|
|
|
|
|
|
#define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
|
2006-09-26 00:31:48 -06:00
|
|
|
#define pmd_offset(dir, address) ((pmd_t *) pud_page_vaddr(*(dir)) + \
|
2005-04-16 16:20:36 -06:00
|
|
|
pmd_index(address))
|
|
|
|
#define pmd_none(x) (!pmd_val(x))
|
|
|
|
#define pmd_present(x) (pmd_val(x) & _PAGE_PRESENT)
|
|
|
|
#define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0)
|
|
|
|
#define pfn_pmd(nr,prot) (__pmd(((nr) << PAGE_SHIFT) | pgprot_val(prot)))
|
2005-11-05 09:25:53 -07:00
|
|
|
#define pmd_pfn(x) ((pmd_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#define pte_to_pgoff(pte) ((pte_val(pte) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
|
|
|
|
#define pgoff_to_pte(off) ((pte_t) { ((off) << PAGE_SHIFT) | _PAGE_FILE })
|
|
|
|
#define PTE_FILE_MAX_BITS __PHYSICAL_MASK_SHIFT
|
|
|
|
|
|
|
|
/* PTE - Level 1 access. */
|
|
|
|
|
|
|
|
/* page, protection -> pte */
|
|
|
|
#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
|
|
|
|
|
|
|
|
#define pte_index(address) \
|
2005-09-30 00:32:19 -06:00
|
|
|
(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
|
2006-09-26 00:31:48 -06:00
|
|
|
#define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
|
2005-04-16 16:20:36 -06:00
|
|
|
pte_index(address))
|
|
|
|
|
|
|
|
/* x86-64 always has all page tables mapped. */
|
|
|
|
#define pte_offset_map(dir,address) pte_offset_kernel(dir,address)
|
|
|
|
#define pte_offset_map_nested(dir,address) pte_offset_kernel(dir,address)
|
|
|
|
#define pte_unmap(pte) /* NOP */
|
|
|
|
#define pte_unmap_nested(pte) /* NOP */
|
|
|
|
|
|
|
|
#define update_mmu_cache(vma,address,pte) do { } while (0)
|
|
|
|
|
|
|
|
/* 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) \
|
2007-06-16 11:16:12 -06:00
|
|
|
({ \
|
|
|
|
int __changed = !pte_same(*(__ptep), __entry); \
|
|
|
|
if (__changed && __dirty) { \
|
|
|
|
set_pte(__ptep, __entry); \
|
|
|
|
flush_tlb_page(__vma, __address); \
|
|
|
|
} \
|
|
|
|
__changed; \
|
|
|
|
})
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* Encode and de-code a swap entry */
|
|
|
|
#define __swp_type(x) (((x).val >> 1) & 0x3f)
|
|
|
|
#define __swp_offset(x) ((x).val >> 8)
|
|
|
|
#define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
|
|
|
|
#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) })
|
|
|
|
#define __swp_entry_to_pte(x) ((pte_t) { (x).val })
|
|
|
|
|
2006-03-25 08:29:40 -07:00
|
|
|
extern spinlock_t pgd_lock;
|
2007-05-02 11:27:10 -06:00
|
|
|
extern struct list_head pgd_list;
|
2006-03-25 08:29:40 -07:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int kern_addr_valid(unsigned long addr);
|
|
|
|
|
2007-07-22 03:12:31 -06:00
|
|
|
pte_t *lookup_address(unsigned long addr);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
#define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \
|
|
|
|
remap_pfn_range(vma, vaddr, pfn, size, prot)
|
|
|
|
|
|
|
|
#define HAVE_ARCH_UNMAPPED_AREA
|
2008-01-30 05:31:07 -07:00
|
|
|
#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#define pgtable_cache_init() do { } while (0)
|
2007-09-21 13:09:41 -06:00
|
|
|
#define check_pgt_cache() do { } while (0)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#define PAGE_AGP PAGE_KERNEL_NOCACHE
|
|
|
|
#define HAVE_PAGE_AGP 1
|
|
|
|
|
|
|
|
/* fs/proc/kcore.c */
|
|
|
|
#define kc_vaddr_to_offset(v) ((v) & __VIRTUAL_MASK)
|
|
|
|
#define kc_offset_to_vaddr(o) \
|
|
|
|
(((o) & (1UL << (__VIRTUAL_MASK_SHIFT-1))) ? ((o) | (~__VIRTUAL_MASK)) : (o))
|
|
|
|
|
|
|
|
#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
|
|
|
|
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
|
2005-09-03 16:55:06 -06:00
|
|
|
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
|
2005-04-16 16:20:36 -06:00
|
|
|
#define __HAVE_ARCH_PTEP_SET_WRPROTECT
|
|
|
|
#define __HAVE_ARCH_PTE_SAME
|
|
|
|
#include <asm-generic/pgtable.h>
|
2007-05-02 11:27:06 -06:00
|
|
|
#endif /* !__ASSEMBLY__ */
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#endif /* _X86_64_PGTABLE_H */
|