2008-10-22 23:26:29 -06:00
|
|
|
#ifndef _ASM_X86_PGTABLE_H
|
|
|
|
#define _ASM_X86_PGTABLE_H
|
2008-01-30 05:32:55 -07:00
|
|
|
|
2009-02-09 03:57:45 -07:00
|
|
|
#include <asm/page.h>
|
|
|
|
|
2009-02-08 19:46:18 -07:00
|
|
|
#include <asm/pgtable_types.h>
|
2008-09-23 15:00:36 -06:00
|
|
|
|
2008-12-18 12:41:31 -07:00
|
|
|
/*
|
|
|
|
* Macro to mark a page protection value as UC-
|
|
|
|
*/
|
|
|
|
#define pgprot_noncached(prot) \
|
|
|
|
((boot_cpu_data.x86 > 3) \
|
|
|
|
? (__pgprot(pgprot_val(prot) | _PAGE_CACHE_UC_MINUS)) \
|
|
|
|
: (prot))
|
|
|
|
|
2008-01-30 05:32:56 -07:00
|
|
|
#ifndef __ASSEMBLY__
|
2008-01-30 05:32:58 -07:00
|
|
|
|
2008-01-30 05:32:58 -07:00
|
|
|
/*
|
|
|
|
* ZERO_PAGE is a global shared page that is always zero: used
|
|
|
|
* for zero-mapped memory areas etc..
|
|
|
|
*/
|
2008-03-23 02:03:12 -06:00
|
|
|
extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
|
2008-01-30 05:32:58 -07:00
|
|
|
#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
|
|
|
|
|
2008-01-30 05:34:11 -07:00
|
|
|
extern spinlock_t pgd_lock;
|
|
|
|
extern struct list_head pgd_list;
|
2008-01-30 05:32:58 -07:00
|
|
|
|
2009-02-11 11:20:05 -07:00
|
|
|
#ifdef CONFIG_PARAVIRT
|
|
|
|
#include <asm/paravirt.h>
|
|
|
|
#else /* !CONFIG_PARAVIRT */
|
|
|
|
#define set_pte(ptep, pte) native_set_pte(ptep, pte)
|
|
|
|
#define set_pte_at(mm, addr, ptep, pte) native_set_pte_at(mm, addr, ptep, pte)
|
|
|
|
|
|
|
|
#define set_pte_atomic(ptep, pte) \
|
|
|
|
native_set_pte_atomic(ptep, pte)
|
|
|
|
|
|
|
|
#define set_pmd(pmdp, pmd) native_set_pmd(pmdp, pmd)
|
|
|
|
|
|
|
|
#ifndef __PAGETABLE_PUD_FOLDED
|
|
|
|
#define set_pgd(pgdp, pgd) native_set_pgd(pgdp, pgd)
|
|
|
|
#define pgd_clear(pgd) native_pgd_clear(pgd)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef set_pud
|
|
|
|
# define set_pud(pudp, pud) native_set_pud(pudp, pud)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef __PAGETABLE_PMD_FOLDED
|
|
|
|
#define pud_clear(pud) native_pud_clear(pud)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#define pte_clear(mm, addr, ptep) native_pte_clear(mm, addr, ptep)
|
|
|
|
#define pmd_clear(pmd) native_pmd_clear(pmd)
|
|
|
|
|
|
|
|
#define pte_update(mm, addr, ptep) do { } while (0)
|
|
|
|
#define pte_update_defer(mm, addr, ptep) do { } while (0)
|
|
|
|
|
|
|
|
static inline void __init paravirt_pagetable_setup_start(pgd_t *base)
|
|
|
|
{
|
|
|
|
native_pagetable_setup_start(base);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void __init paravirt_pagetable_setup_done(pgd_t *base)
|
|
|
|
{
|
|
|
|
native_pagetable_setup_done(base);
|
|
|
|
}
|
|
|
|
|
|
|
|
#define pgd_val(x) native_pgd_val(x)
|
|
|
|
#define __pgd(x) native_make_pgd(x)
|
|
|
|
|
|
|
|
#ifndef __PAGETABLE_PUD_FOLDED
|
|
|
|
#define pud_val(x) native_pud_val(x)
|
|
|
|
#define __pud(x) native_make_pud(x)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef __PAGETABLE_PMD_FOLDED
|
|
|
|
#define pmd_val(x) native_pmd_val(x)
|
|
|
|
#define __pmd(x) native_make_pmd(x)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#define pte_val(x) native_pte_val(x)
|
|
|
|
#define __pte(x) native_make_pte(x)
|
|
|
|
|
|
|
|
#endif /* CONFIG_PARAVIRT */
|
|
|
|
|
2008-01-30 05:32:56 -07:00
|
|
|
/*
|
|
|
|
* The following only work if pte_present() is true.
|
|
|
|
* Undefined behaviour if not..
|
|
|
|
*/
|
2008-03-23 02:03:12 -06:00
|
|
|
static inline int pte_dirty(pte_t pte)
|
|
|
|
{
|
2008-05-26 16:31:06 -06:00
|
|
|
return pte_flags(pte) & _PAGE_DIRTY;
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline int pte_young(pte_t pte)
|
|
|
|
{
|
2008-05-26 16:31:06 -06:00
|
|
|
return pte_flags(pte) & _PAGE_ACCESSED;
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline int pte_write(pte_t pte)
|
|
|
|
{
|
2008-05-26 16:31:06 -06:00
|
|
|
return pte_flags(pte) & _PAGE_RW;
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline int pte_file(pte_t pte)
|
|
|
|
{
|
2008-05-26 16:31:06 -06:00
|
|
|
return pte_flags(pte) & _PAGE_FILE;
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline int pte_huge(pte_t pte)
|
|
|
|
{
|
2008-05-26 16:31:06 -06:00
|
|
|
return pte_flags(pte) & _PAGE_PSE;
|
2008-01-30 05:32:56 -07:00
|
|
|
}
|
|
|
|
|
2008-03-23 02:03:12 -06:00
|
|
|
static inline int pte_global(pte_t pte)
|
|
|
|
{
|
2008-05-26 16:31:06 -06:00
|
|
|
return pte_flags(pte) & _PAGE_GLOBAL;
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline int pte_exec(pte_t pte)
|
|
|
|
{
|
2008-05-26 16:31:06 -06:00
|
|
|
return !(pte_flags(pte) & _PAGE_NX);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
mm: introduce pte_special pte bit
s390 for one, cannot implement VM_MIXEDMAP with pfn_valid, due to their memory
model (which is more dynamic than most). Instead, they had proposed to
implement it with an additional path through vm_normal_page(), using a bit in
the pte to determine whether or not the page should be refcounted:
vm_normal_page()
{
...
if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) {
if (vma->vm_flags & VM_MIXEDMAP) {
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
#else
if (!pfn_valid(pfn))
return NULL;
#endif
goto out;
}
...
}
This is fine, however if we are allowed to use a bit in the pte to determine
refcountedness, we can use that to _completely_ replace all the vma based
schemes. So instead of adding more cases to the already complex vma-based
scheme, we can have a clearly seperate and simple pte-based scheme (and get
slightly better code generation in the process):
vm_normal_page()
{
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
return pte_page(pte);
#else
...
#endif
}
And finally, we may rather make this concept usable by any architecture rather
than making it s390 only, so implement a new type of pte state for this.
Unfortunately the old vma based code must stay, because some architectures may
not be able to spare pte bits. This makes vm_normal_page a little bit more
ugly than we would like, but the 2 cases are clearly seperate.
So introduce a pte_special pte state, and use it in mm/memory.c. It is
currently a noop for all architectures, so this doesn't actually result in any
compiled code changes to mm/memory.o.
BTW:
I haven't put vm_normal_page() into arch code as-per an earlier suggestion.
The reason is that, regardless of where vm_normal_page is actually
implemented, the *abstraction* is still exactly the same. Also, while it
depends on whether the architecture has pte_special or not, that is the
only two possible cases, and it really isn't an arch specific function --
the role of the arch code should be to provide primitive functions and
accessors with which to build the core code; pte_special does that. We do
not want architectures to know or care about vm_normal_page itself, and
we definitely don't want them being able to invent something new there
out of sight of mm/ code. If we made vm_normal_page an arch function, then
we have to make vm_insert_mixed (next patch) an arch function too. So I
don't think moving it to arch code fundamentally improves any abstractions,
while it does practically make the code more difficult to follow, for both
mm and arch developers, and easier to misuse.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Cc: Jared Hulbert <jaredeh@gmail.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-28 03:13:00 -06:00
|
|
|
static inline int pte_special(pte_t pte)
|
|
|
|
{
|
2008-09-17 09:48:17 -06:00
|
|
|
return pte_flags(pte) & _PAGE_SPECIAL;
|
mm: introduce pte_special pte bit
s390 for one, cannot implement VM_MIXEDMAP with pfn_valid, due to their memory
model (which is more dynamic than most). Instead, they had proposed to
implement it with an additional path through vm_normal_page(), using a bit in
the pte to determine whether or not the page should be refcounted:
vm_normal_page()
{
...
if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) {
if (vma->vm_flags & VM_MIXEDMAP) {
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
#else
if (!pfn_valid(pfn))
return NULL;
#endif
goto out;
}
...
}
This is fine, however if we are allowed to use a bit in the pte to determine
refcountedness, we can use that to _completely_ replace all the vma based
schemes. So instead of adding more cases to the already complex vma-based
scheme, we can have a clearly seperate and simple pte-based scheme (and get
slightly better code generation in the process):
vm_normal_page()
{
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
return pte_page(pte);
#else
...
#endif
}
And finally, we may rather make this concept usable by any architecture rather
than making it s390 only, so implement a new type of pte state for this.
Unfortunately the old vma based code must stay, because some architectures may
not be able to spare pte bits. This makes vm_normal_page a little bit more
ugly than we would like, but the 2 cases are clearly seperate.
So introduce a pte_special pte state, and use it in mm/memory.c. It is
currently a noop for all architectures, so this doesn't actually result in any
compiled code changes to mm/memory.o.
BTW:
I haven't put vm_normal_page() into arch code as-per an earlier suggestion.
The reason is that, regardless of where vm_normal_page is actually
implemented, the *abstraction* is still exactly the same. Also, while it
depends on whether the architecture has pte_special or not, that is the
only two possible cases, and it really isn't an arch specific function --
the role of the arch code should be to provide primitive functions and
accessors with which to build the core code; pte_special does that. We do
not want architectures to know or care about vm_normal_page itself, and
we definitely don't want them being able to invent something new there
out of sight of mm/ code. If we made vm_normal_page an arch function, then
we have to make vm_insert_mixed (next patch) an arch function too. So I
don't think moving it to arch code fundamentally improves any abstractions,
while it does practically make the code more difficult to follow, for both
mm and arch developers, and easier to misuse.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Cc: Jared Hulbert <jaredeh@gmail.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-28 03:13:00 -06:00
|
|
|
}
|
|
|
|
|
2008-09-09 09:42:45 -06:00
|
|
|
static inline unsigned long pte_pfn(pte_t pte)
|
|
|
|
{
|
|
|
|
return (pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define pte_page(pte) pfn_to_page(pte_pfn(pte))
|
|
|
|
|
2008-03-23 02:03:12 -06:00
|
|
|
static inline int pmd_large(pmd_t pte)
|
|
|
|
{
|
2009-02-05 12:31:16 -07:00
|
|
|
return (pmd_flags(pte) & (_PAGE_PSE | _PAGE_PRESENT)) ==
|
2008-03-23 02:03:12 -06:00
|
|
|
(_PAGE_PSE | _PAGE_PRESENT);
|
|
|
|
}
|
|
|
|
|
2009-01-22 15:24:22 -07:00
|
|
|
static inline pte_t pte_set_flags(pte_t pte, pteval_t set)
|
|
|
|
{
|
|
|
|
pteval_t v = native_pte_val(pte);
|
|
|
|
|
|
|
|
return native_make_pte(v | set);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_clear_flags(pte_t pte, pteval_t clear)
|
|
|
|
{
|
|
|
|
pteval_t v = native_pte_val(pte);
|
|
|
|
|
|
|
|
return native_make_pte(v & ~clear);
|
|
|
|
}
|
|
|
|
|
2008-03-23 02:03:12 -06:00
|
|
|
static inline pte_t pte_mkclean(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_clear_flags(pte, _PAGE_DIRTY);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_mkold(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_clear_flags(pte, _PAGE_ACCESSED);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_wrprotect(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_clear_flags(pte, _PAGE_RW);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_mkexec(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_clear_flags(pte, _PAGE_NX);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_mkdirty(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_set_flags(pte, _PAGE_DIRTY);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_mkyoung(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_set_flags(pte, _PAGE_ACCESSED);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_mkwrite(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_set_flags(pte, _PAGE_RW);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_mkhuge(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_set_flags(pte, _PAGE_PSE);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_clrhuge(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_clear_flags(pte, _PAGE_PSE);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_mkglobal(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_set_flags(pte, _PAGE_GLOBAL);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pte_t pte_clrglobal(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_clear_flags(pte, _PAGE_GLOBAL);
|
2008-03-23 02:03:12 -06:00
|
|
|
}
|
2008-01-30 05:32:56 -07:00
|
|
|
|
mm: introduce pte_special pte bit
s390 for one, cannot implement VM_MIXEDMAP with pfn_valid, due to their memory
model (which is more dynamic than most). Instead, they had proposed to
implement it with an additional path through vm_normal_page(), using a bit in
the pte to determine whether or not the page should be refcounted:
vm_normal_page()
{
...
if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) {
if (vma->vm_flags & VM_MIXEDMAP) {
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
#else
if (!pfn_valid(pfn))
return NULL;
#endif
goto out;
}
...
}
This is fine, however if we are allowed to use a bit in the pte to determine
refcountedness, we can use that to _completely_ replace all the vma based
schemes. So instead of adding more cases to the already complex vma-based
scheme, we can have a clearly seperate and simple pte-based scheme (and get
slightly better code generation in the process):
vm_normal_page()
{
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
return pte_page(pte);
#else
...
#endif
}
And finally, we may rather make this concept usable by any architecture rather
than making it s390 only, so implement a new type of pte state for this.
Unfortunately the old vma based code must stay, because some architectures may
not be able to spare pte bits. This makes vm_normal_page a little bit more
ugly than we would like, but the 2 cases are clearly seperate.
So introduce a pte_special pte state, and use it in mm/memory.c. It is
currently a noop for all architectures, so this doesn't actually result in any
compiled code changes to mm/memory.o.
BTW:
I haven't put vm_normal_page() into arch code as-per an earlier suggestion.
The reason is that, regardless of where vm_normal_page is actually
implemented, the *abstraction* is still exactly the same. Also, while it
depends on whether the architecture has pte_special or not, that is the
only two possible cases, and it really isn't an arch specific function --
the role of the arch code should be to provide primitive functions and
accessors with which to build the core code; pte_special does that. We do
not want architectures to know or care about vm_normal_page itself, and
we definitely don't want them being able to invent something new there
out of sight of mm/ code. If we made vm_normal_page an arch function, then
we have to make vm_insert_mixed (next patch) an arch function too. So I
don't think moving it to arch code fundamentally improves any abstractions,
while it does practically make the code more difficult to follow, for both
mm and arch developers, and easier to misuse.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Cc: Jared Hulbert <jaredeh@gmail.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-28 03:13:00 -06:00
|
|
|
static inline pte_t pte_mkspecial(pte_t pte)
|
|
|
|
{
|
2009-01-22 15:24:22 -07:00
|
|
|
return pte_set_flags(pte, _PAGE_SPECIAL);
|
mm: introduce pte_special pte bit
s390 for one, cannot implement VM_MIXEDMAP with pfn_valid, due to their memory
model (which is more dynamic than most). Instead, they had proposed to
implement it with an additional path through vm_normal_page(), using a bit in
the pte to determine whether or not the page should be refcounted:
vm_normal_page()
{
...
if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) {
if (vma->vm_flags & VM_MIXEDMAP) {
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
#else
if (!pfn_valid(pfn))
return NULL;
#endif
goto out;
}
...
}
This is fine, however if we are allowed to use a bit in the pte to determine
refcountedness, we can use that to _completely_ replace all the vma based
schemes. So instead of adding more cases to the already complex vma-based
scheme, we can have a clearly seperate and simple pte-based scheme (and get
slightly better code generation in the process):
vm_normal_page()
{
#ifdef s390
if (!mixedmap_refcount_pte(pte))
return NULL;
return pte_page(pte);
#else
...
#endif
}
And finally, we may rather make this concept usable by any architecture rather
than making it s390 only, so implement a new type of pte state for this.
Unfortunately the old vma based code must stay, because some architectures may
not be able to spare pte bits. This makes vm_normal_page a little bit more
ugly than we would like, but the 2 cases are clearly seperate.
So introduce a pte_special pte state, and use it in mm/memory.c. It is
currently a noop for all architectures, so this doesn't actually result in any
compiled code changes to mm/memory.o.
BTW:
I haven't put vm_normal_page() into arch code as-per an earlier suggestion.
The reason is that, regardless of where vm_normal_page is actually
implemented, the *abstraction* is still exactly the same. Also, while it
depends on whether the architecture has pte_special or not, that is the
only two possible cases, and it really isn't an arch specific function --
the role of the arch code should be to provide primitive functions and
accessors with which to build the core code; pte_special does that. We do
not want architectures to know or care about vm_normal_page itself, and
we definitely don't want them being able to invent something new there
out of sight of mm/ code. If we made vm_normal_page an arch function, then
we have to make vm_insert_mixed (next patch) an arch function too. So I
don't think moving it to arch code fundamentally improves any abstractions,
while it does practically make the code more difficult to follow, for both
mm and arch developers, and easier to misuse.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Cc: Jared Hulbert <jaredeh@gmail.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-28 03:13:00 -06:00
|
|
|
}
|
|
|
|
|
2009-02-04 19:33:38 -07:00
|
|
|
/*
|
|
|
|
* Mask out unsupported bits in a present pgprot. Non-present pgprots
|
|
|
|
* can use those bits for other purposes, so leave them be.
|
|
|
|
*/
|
|
|
|
static inline pgprotval_t massage_pgprot(pgprot_t pgprot)
|
|
|
|
{
|
|
|
|
pgprotval_t protval = pgprot_val(pgprot);
|
|
|
|
|
|
|
|
if (protval & _PAGE_PRESENT)
|
|
|
|
protval &= __supported_pte_mask;
|
|
|
|
|
|
|
|
return protval;
|
|
|
|
}
|
|
|
|
|
2008-01-30 05:32:57 -07:00
|
|
|
static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
|
|
|
|
{
|
2009-02-04 19:33:38 -07:00
|
|
|
return __pte(((phys_addr_t)page_nr << PAGE_SHIFT) |
|
|
|
|
massage_pgprot(pgprot));
|
2008-01-30 05:32:57 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
|
|
|
|
{
|
2009-02-04 19:33:38 -07:00
|
|
|
return __pmd(((phys_addr_t)page_nr << PAGE_SHIFT) |
|
|
|
|
massage_pgprot(pgprot));
|
2008-01-30 05:32:57 -07:00
|
|
|
}
|
|
|
|
|
2008-01-30 05:32:57 -07:00
|
|
|
static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
|
|
|
|
{
|
|
|
|
pteval_t val = pte_val(pte);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Chop off the NX bit (if present), and add the NX portion of
|
|
|
|
* the newprot (if present):
|
|
|
|
*/
|
2008-05-14 17:05:51 -06:00
|
|
|
val &= _PAGE_CHG_MASK;
|
2009-02-04 19:33:38 -07:00
|
|
|
val |= massage_pgprot(newprot) & ~_PAGE_CHG_MASK;
|
2008-01-30 05:32:57 -07:00
|
|
|
|
|
|
|
return __pte(val);
|
|
|
|
}
|
|
|
|
|
2008-05-14 17:05:51 -06:00
|
|
|
/* mprotect needs to preserve PAT bits when updating vm_page_prot */
|
|
|
|
#define pgprot_modify pgprot_modify
|
|
|
|
static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot)
|
|
|
|
{
|
|
|
|
pgprotval_t preservebits = pgprot_val(oldprot) & _PAGE_CHG_MASK;
|
|
|
|
pgprotval_t addbits = pgprot_val(newprot);
|
|
|
|
return __pgprot(preservebits | addbits);
|
|
|
|
}
|
|
|
|
|
2008-07-21 23:59:56 -06:00
|
|
|
#define pte_pgprot(x) __pgprot(pte_flags(x) & PTE_FLAGS_MASK)
|
2008-01-30 05:33:51 -07:00
|
|
|
|
2009-02-04 19:33:38 -07:00
|
|
|
#define canon_pgprot(p) __pgprot(massage_pgprot(p))
|
2008-01-30 05:33:53 -07:00
|
|
|
|
2009-01-09 17:13:10 -07:00
|
|
|
static inline int is_new_memtype_allowed(unsigned long flags,
|
|
|
|
unsigned long new_flags)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* Certain new memtypes are not allowed with certain
|
|
|
|
* requested memtype:
|
|
|
|
* - request is uncached, return cannot be write-back
|
|
|
|
* - request is write-combine, return cannot be write-back
|
|
|
|
*/
|
|
|
|
if ((flags == _PAGE_CACHE_UC_MINUS &&
|
|
|
|
new_flags == _PAGE_CACHE_WB) ||
|
|
|
|
(flags == _PAGE_CACHE_WC &&
|
|
|
|
new_flags == _PAGE_CACHE_WB)) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-02-23 19:57:21 -07:00
|
|
|
pmd_t *populate_extra_pmd(unsigned long vaddr);
|
|
|
|
pte_t *populate_extra_pte(unsigned long vaddr);
|
2008-01-30 05:32:56 -07:00
|
|
|
#endif /* __ASSEMBLY__ */
|
|
|
|
|
2007-10-11 03:20:03 -06:00
|
|
|
#ifdef CONFIG_X86_32
|
|
|
|
# include "pgtable_32.h"
|
|
|
|
#else
|
|
|
|
# include "pgtable_64.h"
|
|
|
|
#endif
|
2008-01-30 05:32:55 -07:00
|
|
|
|
2009-02-05 12:30:54 -07:00
|
|
|
#ifndef __ASSEMBLY__
|
2009-02-05 12:30:55 -07:00
|
|
|
#include <linux/mm_types.h>
|
2009-02-05 12:30:54 -07:00
|
|
|
|
2009-02-05 12:30:43 -07:00
|
|
|
static inline int pte_none(pte_t pte)
|
|
|
|
{
|
|
|
|
return !pte.pte;
|
|
|
|
}
|
|
|
|
|
2009-02-05 12:30:44 -07:00
|
|
|
#define __HAVE_ARCH_PTE_SAME
|
|
|
|
static inline int pte_same(pte_t a, pte_t b)
|
|
|
|
{
|
|
|
|
return a.pte == b.pte;
|
|
|
|
}
|
|
|
|
|
2009-02-05 12:30:45 -07:00
|
|
|
static inline int pte_present(pte_t a)
|
|
|
|
{
|
|
|
|
return pte_flags(a) & (_PAGE_PRESENT | _PAGE_PROTNONE);
|
|
|
|
}
|
|
|
|
|
2009-02-05 12:30:50 -07:00
|
|
|
static inline int pmd_present(pmd_t pmd)
|
|
|
|
{
|
2009-02-05 12:31:16 -07:00
|
|
|
return pmd_flags(pmd) & _PAGE_PRESENT;
|
2009-02-05 12:30:50 -07:00
|
|
|
}
|
|
|
|
|
2009-02-05 12:30:51 -07:00
|
|
|
static inline int pmd_none(pmd_t pmd)
|
|
|
|
{
|
|
|
|
/* Only check low word on 32-bit platforms, since it might be
|
|
|
|
out of sync with upper half. */
|
2009-02-05 12:31:17 -07:00
|
|
|
return (unsigned long)native_pmd_val(pmd) == 0;
|
2009-02-05 12:30:51 -07:00
|
|
|
}
|
|
|
|
|
2009-02-05 12:30:59 -07:00
|
|
|
static inline unsigned long pmd_page_vaddr(pmd_t pmd)
|
|
|
|
{
|
|
|
|
return (unsigned long)__va(pmd_val(pmd) & PTE_PFN_MASK);
|
|
|
|
}
|
|
|
|
|
2009-02-09 03:42:57 -07:00
|
|
|
/*
|
|
|
|
* Currently stuck as a macro due to indirect forward reference to
|
|
|
|
* linux/mmzone.h's __section_mem_map_addr() definition:
|
|
|
|
*/
|
|
|
|
#define pmd_page(pmd) pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT)
|
2009-02-05 12:31:00 -07:00
|
|
|
|
2009-02-05 12:31:01 -07:00
|
|
|
/*
|
|
|
|
* the pmd page can be thought of an array like this: pmd_t[PTRS_PER_PMD]
|
|
|
|
*
|
|
|
|
* this macro returns the index of the entry in the pmd page which would
|
|
|
|
* control the given virtual address
|
|
|
|
*/
|
|
|
|
static inline unsigned pmd_index(unsigned long address)
|
|
|
|
{
|
|
|
|
return (address >> PMD_SHIFT) & (PTRS_PER_PMD - 1);
|
|
|
|
}
|
|
|
|
|
2009-02-05 12:31:05 -07:00
|
|
|
/*
|
|
|
|
* Conversion functions: convert a page and protection to a page entry,
|
|
|
|
* and a page entry and page directory to the page they refer to.
|
|
|
|
*
|
|
|
|
* (Currently stuck as a macro because of indirect forward reference
|
|
|
|
* to linux/mm.h:page_to_nid())
|
|
|
|
*/
|
|
|
|
#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
|
|
|
|
|
2009-02-05 12:31:06 -07:00
|
|
|
/*
|
|
|
|
* the pte page can be thought of an array like this: pte_t[PTRS_PER_PTE]
|
|
|
|
*
|
|
|
|
* this function returns the index of the entry in the pte page which would
|
|
|
|
* control the given virtual address
|
|
|
|
*/
|
|
|
|
static inline unsigned pte_index(unsigned long address)
|
|
|
|
{
|
|
|
|
return (address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1);
|
|
|
|
}
|
|
|
|
|
2009-02-05 12:31:07 -07:00
|
|
|
static inline pte_t *pte_offset_kernel(pmd_t *pmd, unsigned long address)
|
|
|
|
{
|
|
|
|
return (pte_t *)pmd_page_vaddr(*pmd) + pte_index(address);
|
|
|
|
}
|
|
|
|
|
2009-02-05 12:31:11 -07:00
|
|
|
static inline int pmd_bad(pmd_t pmd)
|
|
|
|
{
|
2009-02-05 12:31:16 -07:00
|
|
|
return (pmd_flags(pmd) & ~_PAGE_USER) != _KERNPG_TABLE;
|
2009-02-05 12:31:11 -07:00
|
|
|
}
|
|
|
|
|
2009-02-05 12:31:12 -07:00
|
|
|
static inline unsigned long pages_to_mb(unsigned long npg)
|
|
|
|
{
|
|
|
|
return npg >> (20 - PAGE_SHIFT);
|
|
|
|
}
|
|
|
|
|
2009-02-05 12:31:15 -07:00
|
|
|
#define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \
|
|
|
|
remap_pfn_range(vma, vaddr, pfn, size, prot)
|
|
|
|
|
2009-02-05 12:30:48 -07:00
|
|
|
#if PAGETABLE_LEVELS > 2
|
2009-02-05 12:31:13 -07:00
|
|
|
static inline int pud_none(pud_t pud)
|
|
|
|
{
|
2009-02-05 12:31:17 -07:00
|
|
|
return native_pud_val(pud) == 0;
|
2009-02-05 12:31:13 -07:00
|
|
|
}
|
|
|
|
|
2009-02-05 12:30:48 -07:00
|
|
|
static inline int pud_present(pud_t pud)
|
|
|
|
{
|
2009-02-05 12:31:16 -07:00
|
|
|
return pud_flags(pud) & _PAGE_PRESENT;
|
2009-02-05 12:30:48 -07:00
|
|
|
}
|
2009-02-05 12:30:53 -07:00
|
|
|
|
|
|
|
static inline unsigned long pud_page_vaddr(pud_t pud)
|
|
|
|
{
|
|
|
|
return (unsigned long)__va((unsigned long)pud_val(pud) & PTE_PFN_MASK);
|
|
|
|
}
|
2009-02-05 12:30:55 -07:00
|
|
|
|
2009-02-09 03:42:57 -07:00
|
|
|
/*
|
|
|
|
* Currently stuck as a macro due to indirect forward reference to
|
|
|
|
* linux/mmzone.h's __section_mem_map_addr() definition:
|
|
|
|
*/
|
|
|
|
#define pud_page(pud) pfn_to_page(pud_val(pud) >> PAGE_SHIFT)
|
2009-02-05 12:31:02 -07:00
|
|
|
|
|
|
|
/* Find an entry in the second-level page table.. */
|
|
|
|
static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address)
|
|
|
|
{
|
|
|
|
return (pmd_t *)pud_page_vaddr(*pud) + pmd_index(address);
|
|
|
|
}
|
2009-02-05 12:31:04 -07:00
|
|
|
|
|
|
|
static inline unsigned long pmd_pfn(pmd_t pmd)
|
|
|
|
{
|
|
|
|
return (pmd_val(pmd) & PTE_PFN_MASK) >> PAGE_SHIFT;
|
|
|
|
}
|
2009-02-05 12:31:08 -07:00
|
|
|
|
|
|
|
static inline int pud_large(pud_t pud)
|
|
|
|
{
|
2009-02-09 01:09:52 -07:00
|
|
|
return (pud_val(pud) & (_PAGE_PSE | _PAGE_PRESENT)) ==
|
2009-02-05 12:31:08 -07:00
|
|
|
(_PAGE_PSE | _PAGE_PRESENT);
|
|
|
|
}
|
2009-02-05 12:31:10 -07:00
|
|
|
|
|
|
|
static inline int pud_bad(pud_t pud)
|
|
|
|
{
|
2009-02-05 12:31:16 -07:00
|
|
|
return (pud_flags(pud) & ~(_KERNPG_TABLE | _PAGE_USER)) != 0;
|
2009-02-05 12:31:10 -07:00
|
|
|
}
|
2009-02-09 01:09:52 -07:00
|
|
|
#else
|
|
|
|
static inline int pud_large(pud_t pud)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
2009-02-05 12:30:48 -07:00
|
|
|
#endif /* PAGETABLE_LEVELS > 2 */
|
|
|
|
|
2009-02-05 12:30:49 -07:00
|
|
|
#if PAGETABLE_LEVELS > 3
|
|
|
|
static inline int pgd_present(pgd_t pgd)
|
|
|
|
{
|
2009-02-05 12:31:16 -07:00
|
|
|
return pgd_flags(pgd) & _PAGE_PRESENT;
|
2009-02-05 12:30:49 -07:00
|
|
|
}
|
2009-02-05 12:30:52 -07:00
|
|
|
|
|
|
|
static inline unsigned long pgd_page_vaddr(pgd_t pgd)
|
|
|
|
{
|
|
|
|
return (unsigned long)__va((unsigned long)pgd_val(pgd) & PTE_PFN_MASK);
|
|
|
|
}
|
2009-02-05 12:30:56 -07:00
|
|
|
|
2009-02-09 03:42:57 -07:00
|
|
|
/*
|
|
|
|
* Currently stuck as a macro due to indirect forward reference to
|
|
|
|
* linux/mmzone.h's __section_mem_map_addr() definition:
|
|
|
|
*/
|
|
|
|
#define pgd_page(pgd) pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT)
|
2009-02-05 12:30:57 -07:00
|
|
|
|
|
|
|
/* to find an entry in a page-table-directory. */
|
|
|
|
static inline unsigned pud_index(unsigned long address)
|
|
|
|
{
|
|
|
|
return (address >> PUD_SHIFT) & (PTRS_PER_PUD - 1);
|
|
|
|
}
|
2009-02-05 12:30:58 -07:00
|
|
|
|
|
|
|
static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address)
|
|
|
|
{
|
|
|
|
return (pud_t *)pgd_page_vaddr(*pgd) + pud_index(address);
|
|
|
|
}
|
2009-02-05 12:31:09 -07:00
|
|
|
|
|
|
|
static inline int pgd_bad(pgd_t pgd)
|
|
|
|
{
|
2009-02-05 12:31:16 -07:00
|
|
|
return (pgd_flags(pgd) & ~_PAGE_USER) != _KERNPG_TABLE;
|
2009-02-05 12:31:09 -07:00
|
|
|
}
|
2009-02-05 12:31:14 -07:00
|
|
|
|
|
|
|
static inline int pgd_none(pgd_t pgd)
|
|
|
|
{
|
2009-02-05 12:31:17 -07:00
|
|
|
return !native_pgd_val(pgd);
|
2009-02-05 12:31:14 -07:00
|
|
|
}
|
2009-02-05 12:30:49 -07:00
|
|
|
#endif /* PAGETABLE_LEVELS > 3 */
|
|
|
|
|
2008-01-30 05:32:56 -07:00
|
|
|
#endif /* __ASSEMBLY__ */
|
|
|
|
|
2008-06-24 22:19:06 -06:00
|
|
|
/*
|
|
|
|
* the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
|
|
|
|
*
|
|
|
|
* this macro returns the index of the entry in the pgd page which would
|
|
|
|
* control the given virtual address
|
|
|
|
*/
|
|
|
|
#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))
|
|
|
|
|
|
|
|
/*
|
|
|
|
* pgd_offset() returns a (pgd_t *)
|
|
|
|
* pgd_index() is used get the offset into the pgd page's array of pgd_t's;
|
|
|
|
*/
|
|
|
|
#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address)))
|
|
|
|
/*
|
|
|
|
* a shortcut which implies the use of the kernel's pgd, instead
|
|
|
|
* of a process's
|
|
|
|
*/
|
|
|
|
#define pgd_offset_k(address) pgd_offset(&init_mm, (address))
|
|
|
|
|
|
|
|
|
2008-03-17 17:37:13 -06:00
|
|
|
#define KERNEL_PGD_BOUNDARY pgd_index(PAGE_OFFSET)
|
|
|
|
#define KERNEL_PGD_PTRS (PTRS_PER_PGD - KERNEL_PGD_BOUNDARY)
|
|
|
|
|
2008-01-30 05:32:58 -07:00
|
|
|
#ifndef __ASSEMBLY__
|
|
|
|
|
2008-01-30 05:32:58 -07:00
|
|
|
/* local pte updates need not use xchg for locking */
|
|
|
|
static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep)
|
|
|
|
{
|
|
|
|
pte_t res = *ptep;
|
|
|
|
|
|
|
|
/* Pure native function needs no input for mm, addr */
|
|
|
|
native_pte_clear(NULL, 0, ptep);
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
|
|
|
|
pte_t *ptep , pte_t pte)
|
|
|
|
{
|
|
|
|
native_set_pte(ptep, pte);
|
|
|
|
}
|
|
|
|
|
2008-01-30 05:32:58 -07:00
|
|
|
#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.
|
|
|
|
*/
|
2008-06-24 22:18:57 -06:00
|
|
|
struct vm_area_struct;
|
|
|
|
|
2008-01-30 05:32:58 -07:00
|
|
|
#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
|
2008-03-17 17:37:03 -06:00
|
|
|
extern int ptep_set_access_flags(struct vm_area_struct *vma,
|
|
|
|
unsigned long address, pte_t *ptep,
|
|
|
|
pte_t entry, int dirty);
|
2008-01-30 05:32:58 -07:00
|
|
|
|
|
|
|
#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
|
2008-03-17 17:37:04 -06:00
|
|
|
extern int ptep_test_and_clear_young(struct vm_area_struct *vma,
|
|
|
|
unsigned long addr, pte_t *ptep);
|
2008-01-30 05:32:58 -07:00
|
|
|
|
|
|
|
#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
|
2008-03-17 17:37:05 -06:00
|
|
|
extern int ptep_clear_flush_young(struct vm_area_struct *vma,
|
|
|
|
unsigned long address, pte_t *ptep);
|
2008-01-30 05:32:58 -07:00
|
|
|
|
|
|
|
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
|
2008-03-23 02:03:12 -06:00
|
|
|
static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
|
|
|
|
pte_t *ptep)
|
2008-01-30 05:32:58 -07:00
|
|
|
{
|
|
|
|
pte_t pte = native_ptep_get_and_clear(ptep);
|
|
|
|
pte_update(mm, addr, ptep);
|
|
|
|
return pte;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
|
2008-03-23 02:03:12 -06:00
|
|
|
static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
|
|
|
|
unsigned long addr, pte_t *ptep,
|
|
|
|
int full)
|
2008-01-30 05:32:58 -07:00
|
|
|
{
|
|
|
|
pte_t pte;
|
|
|
|
if (full) {
|
|
|
|
/*
|
|
|
|
* Full address destruction in progress; paravirt does not
|
|
|
|
* care about updates and native needs no locking
|
|
|
|
*/
|
|
|
|
pte = native_local_ptep_get_and_clear(ptep);
|
|
|
|
} else {
|
|
|
|
pte = ptep_get_and_clear(mm, addr, ptep);
|
|
|
|
}
|
|
|
|
return pte;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define __HAVE_ARCH_PTEP_SET_WRPROTECT
|
2008-03-23 02:03:12 -06:00
|
|
|
static inline void ptep_set_wrprotect(struct mm_struct *mm,
|
|
|
|
unsigned long addr, pte_t *ptep)
|
2008-01-30 05:32:58 -07:00
|
|
|
{
|
2008-01-30 05:32:58 -07:00
|
|
|
clear_bit(_PAGE_BIT_RW, (unsigned long *)&ptep->pte);
|
2008-01-30 05:32:58 -07:00
|
|
|
pte_update(mm, addr, ptep);
|
|
|
|
}
|
|
|
|
|
2008-03-17 17:37:14 -06:00
|
|
|
/*
|
|
|
|
* clone_pgd_range(pgd_t *dst, pgd_t *src, int count);
|
|
|
|
*
|
|
|
|
* dst - pointer to pgd range anwhere on a pgd page
|
|
|
|
* src - ""
|
|
|
|
* count - the number of pgds to copy.
|
|
|
|
*
|
|
|
|
* dst and src can be on the same page, but the range must not overlap,
|
|
|
|
* and must not cross a page boundary.
|
|
|
|
*/
|
|
|
|
static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
|
|
|
|
{
|
|
|
|
memcpy(dst, src, count * sizeof(pgd_t));
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2008-01-30 05:32:58 -07:00
|
|
|
#include <asm-generic/pgtable.h>
|
|
|
|
#endif /* __ASSEMBLY__ */
|
|
|
|
|
2008-10-22 23:26:29 -06:00
|
|
|
#endif /* _ASM_X86_PGTABLE_H */
|