[deliverable/linux.git] / include / asm-powerpc / pgtable-ppc64.h

#ifndef _ASM_POWERPC_PGTABLE_PPC64_H_
#define _ASM_POWERPC_PGTABLE_PPC64_H_
/*
 * This file contains the functions and defines necessary to modify and use
 * the ppc64 hashed page table.
 */

#ifndef __ASSEMBLY__
#include <linux/stddef.h>
#include <asm/tlbflush.h>
#endif /* __ASSEMBLY__ */

#ifdef CONFIG_PPC_64K_PAGES
#include <asm/pgtable-64k.h>
#else
#include <asm/pgtable-4k.h>
#endif

#define FIRST_USER_ADDRESS	0

/*
 * Size of EA range mapped by our pagetables.
 */
#define PGTABLE_EADDR_SIZE (PTE_INDEX_SIZE + PMD_INDEX_SIZE + \
                	    PUD_INDEX_SIZE + PGD_INDEX_SIZE + PAGE_SHIFT)
#define PGTABLE_RANGE (ASM_CONST(1) << PGTABLE_EADDR_SIZE)

#if TASK_SIZE_USER64 > PGTABLE_RANGE
#error TASK_SIZE_USER64 exceeds pagetable range
#endif

#if TASK_SIZE_USER64 > (1UL << (USER_ESID_BITS + SID_SHIFT))
#error TASK_SIZE_USER64 exceeds user VSID range
#endif


/*
 * Define the address range of the vmalloc VM area.
 */
#define VMALLOC_START ASM_CONST(0xD000000000000000)
#define VMALLOC_SIZE  (PGTABLE_RANGE >> 1)
#define VMALLOC_END   (VMALLOC_START + VMALLOC_SIZE)

/*
 * Define the address ranges for MMIO and IO space :
 *
 *  ISA_IO_BASE = VMALLOC_END, 64K reserved area
 *  PHB_IO_BASE = ISA_IO_BASE + 64K to ISA_IO_BASE + 2G, PHB IO spaces
 * IOREMAP_BASE = ISA_IO_BASE + 2G to VMALLOC_START + PGTABLE_RANGE
 */
#define FULL_IO_SIZE	0x80000000ul
#define  ISA_IO_BASE	(VMALLOC_END)
#define  ISA_IO_END	(VMALLOC_END + 0x10000ul)
#define  PHB_IO_BASE	(ISA_IO_END)
#define  PHB_IO_END	(VMALLOC_END + FULL_IO_SIZE)
#define IOREMAP_BASE	(PHB_IO_END)
#define IOREMAP_END	(VMALLOC_START + PGTABLE_RANGE)

/*
 * Region IDs
 */
#define REGION_SHIFT		60UL
#define REGION_MASK		(0xfUL << REGION_SHIFT)
#define REGION_ID(ea)		(((unsigned long)(ea)) >> REGION_SHIFT)

#define VMALLOC_REGION_ID	(REGION_ID(VMALLOC_START))
#define KERNEL_REGION_ID	(REGION_ID(PAGE_OFFSET))
#define VMEMMAP_REGION_ID	(0xfUL)
#define USER_REGION_ID		(0UL)

/*
 * Defines the address of the vmemap area, in its own region
 */
#define VMEMMAP_BASE		(VMEMMAP_REGION_ID << REGION_SHIFT)
#define vmemmap			((struct page *)VMEMMAP_BASE)


/*
 * Common bits in a linux-style PTE.  These match the bits in the
 * (hardware-defined) PowerPC PTE as closely as possible. Additional
 * bits may be defined in pgtable-*.h
 */
#define _PAGE_PRESENT	0x0001 /* software: pte contains a translation */
#define _PAGE_USER	0x0002 /* matches one of the PP bits */
#define _PAGE_FILE	0x0002 /* (!present only) software: pte holds file offset */
#define _PAGE_EXEC	0x0004 /* No execute on POWER4 and newer (we invert) */
#define _PAGE_GUARDED	0x0008
#define _PAGE_COHERENT	0x0010 /* M: enforce memory coherence (SMP systems) */
#define _PAGE_NO_CACHE	0x0020 /* I: cache inhibit */
#define _PAGE_WRITETHRU	0x0040 /* W: cache write-through */
#define _PAGE_DIRTY	0x0080 /* C: page changed */
#define _PAGE_ACCESSED	0x0100 /* R: page referenced */
#define _PAGE_RW	0x0200 /* software: user write access allowed */
#define _PAGE_HASHPTE	0x0400 /* software: pte has an associated HPTE */
#define _PAGE_BUSY	0x0800 /* software: PTE & hash are busy */

#define _PAGE_BASE	(_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_COHERENT)

#define _PAGE_WRENABLE	(_PAGE_RW | _PAGE_DIRTY)

/* __pgprot defined in asm-powerpc/page.h */
#define PAGE_NONE	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED)

#define PAGE_SHARED	__pgprot(_PAGE_BASE | _PAGE_RW | _PAGE_USER)
#define PAGE_SHARED_X	__pgprot(_PAGE_BASE | _PAGE_RW | _PAGE_USER | _PAGE_EXEC)
#define PAGE_COPY	__pgprot(_PAGE_BASE | _PAGE_USER)
#define PAGE_COPY_X	__pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC)
#define PAGE_READONLY	__pgprot(_PAGE_BASE | _PAGE_USER)
#define PAGE_READONLY_X	__pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC)
#define PAGE_KERNEL	__pgprot(_PAGE_BASE | _PAGE_WRENABLE)
#define PAGE_KERNEL_CI	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
			       _PAGE_WRENABLE | _PAGE_NO_CACHE | _PAGE_GUARDED)
#define PAGE_KERNEL_EXEC __pgprot(_PAGE_BASE | _PAGE_WRENABLE | _PAGE_EXEC)

#define PAGE_AGP	__pgprot(_PAGE_BASE | _PAGE_WRENABLE | _PAGE_NO_CACHE)
#define HAVE_PAGE_AGP

/* PTEIDX nibble */
#define _PTEIDX_SECONDARY	0x8
#define _PTEIDX_GROUP_IX	0x7


/*
 * POWER4 and newer have per page execute protection, older chips can only
 * do this on a segment (256MB) basis.
 *
 * Also, write permissions imply read permissions.
 * This is the closest we can get..
 *
 * Note due to the way vm flags are laid out, the bits are XWR
 */
#define __P000	PAGE_NONE
#define __P001	PAGE_READONLY
#define __P010	PAGE_COPY
#define __P011	PAGE_COPY
#define __P100	PAGE_READONLY_X
#define __P101	PAGE_READONLY_X
#define __P110	PAGE_COPY_X
#define __P111	PAGE_COPY_X

#define __S000	PAGE_NONE
#define __S001	PAGE_READONLY
#define __S010	PAGE_SHARED
#define __S011	PAGE_SHARED
#define __S100	PAGE_READONLY_X
#define __S101	PAGE_READONLY_X
#define __S110	PAGE_SHARED_X
#define __S111	PAGE_SHARED_X

#ifdef CONFIG_HUGETLB_PAGE

#define HAVE_ARCH_UNMAPPED_AREA
#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN

#endif

#ifndef __ASSEMBLY__

/*
 * Conversion functions: convert a page and protection to a page entry,
 * and a page entry and page directory to the page they refer to.
 *
 * mk_pte takes a (struct page *) as input
 */
#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))

static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
{
	pte_t pte;


	pte_val(pte) = (pfn << PTE_RPN_SHIFT) | pgprot_val(pgprot);
	return pte;
}

#define pte_modify(_pte, newprot) \
  (__pte((pte_val(_pte) & _PAGE_CHG_MASK) | pgprot_val(newprot)))

#define pte_none(pte)		((pte_val(pte) & ~_PAGE_HPTEFLAGS) == 0)
#define pte_present(pte)	(pte_val(pte) & _PAGE_PRESENT)

/* pte_clear moved to later in this file */

#define pte_pfn(x)		((unsigned long)((pte_val(x)>>PTE_RPN_SHIFT)))
#define pte_page(x)		pfn_to_page(pte_pfn(x))

#define PMD_BAD_BITS		(PTE_TABLE_SIZE-1)
#define PUD_BAD_BITS		(PMD_TABLE_SIZE-1)

#define pmd_set(pmdp, pmdval) 	(pmd_val(*(pmdp)) = (pmdval))
#define pmd_none(pmd)		(!pmd_val(pmd))
#define	pmd_bad(pmd)		(!is_kernel_addr(pmd_val(pmd)) \
				 || (pmd_val(pmd) & PMD_BAD_BITS))
#define	pmd_present(pmd)	(pmd_val(pmd) != 0)
#define	pmd_clear(pmdp)		(pmd_val(*(pmdp)) = 0)
#define pmd_page_vaddr(pmd)	(pmd_val(pmd) & ~PMD_MASKED_BITS)
#define pmd_page(pmd)		virt_to_page(pmd_page_vaddr(pmd))

#define pud_set(pudp, pudval)	(pud_val(*(pudp)) = (pudval))
#define pud_none(pud)		(!pud_val(pud))
#define	pud_bad(pud)		(!is_kernel_addr(pud_val(pud)) \
				 || (pud_val(pud) & PUD_BAD_BITS))
#define pud_present(pud)	(pud_val(pud) != 0)
#define pud_clear(pudp)		(pud_val(*(pudp)) = 0)
#define pud_page_vaddr(pud)	(pud_val(pud) & ~PUD_MASKED_BITS)
#define pud_page(pud)		virt_to_page(pud_page_vaddr(pud))

#define pgd_set(pgdp, pudp)	({pgd_val(*(pgdp)) = (unsigned long)(pudp);})

/*
 * Find an entry in a page-table-directory.  We combine the address region
 * (the high order N bits) and the pgd portion of the address.
 */
/* to avoid overflow in free_pgtables we don't use PTRS_PER_PGD here */
#define pgd_index(address) (((address) >> (PGDIR_SHIFT)) & 0x1ff)

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

#define pmd_offset(pudp,addr) \
  (((pmd_t *) pud_page_vaddr(*(pudp))) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))

#define pte_offset_kernel(dir,addr) \
  (((pte_t *) pmd_page_vaddr(*(dir))) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)))

#define pte_offset_map(dir,addr)	pte_offset_kernel((dir), (addr))
#define pte_offset_map_nested(dir,addr)	pte_offset_kernel((dir), (addr))
#define pte_unmap(pte)			do { } while(0)
#define pte_unmap_nested(pte)		do { } while(0)

/* to find an entry in a kernel page-table-directory */
/* This now only contains the vmalloc pages */
#define pgd_offset_k(address) pgd_offset(&init_mm, address)

/*
 * The following only work if pte_present() is true.
 * Undefined behaviour if not..
 */
static inline int pte_write(pte_t pte) { return pte_val(pte) & _PAGE_RW;}
static inline int pte_dirty(pte_t pte) { return pte_val(pte) & _PAGE_DIRTY;}
static inline int pte_young(pte_t pte) { return pte_val(pte) & _PAGE_ACCESSED;}
static inline int pte_file(pte_t pte) { return pte_val(pte) & _PAGE_FILE;}
static inline int pte_special(pte_t pte) { return 0; }

static inline void pte_uncache(pte_t pte) { pte_val(pte) |= _PAGE_NO_CACHE; }
static inline void pte_cache(pte_t pte)   { pte_val(pte) &= ~_PAGE_NO_CACHE; }

static inline pte_t pte_wrprotect(pte_t pte) {
	pte_val(pte) &= ~(_PAGE_RW); return pte; }
static inline pte_t pte_mkclean(pte_t pte) {
	pte_val(pte) &= ~(_PAGE_DIRTY); return pte; }
static inline pte_t pte_mkold(pte_t pte) {
	pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
static inline pte_t pte_mkwrite(pte_t pte) {
	pte_val(pte) |= _PAGE_RW; return pte; }
static inline pte_t pte_mkdirty(pte_t pte) {
	pte_val(pte) |= _PAGE_DIRTY; return pte; }
static inline pte_t pte_mkyoung(pte_t pte) {
	pte_val(pte) |= _PAGE_ACCESSED; return pte; }
static inline pte_t pte_mkhuge(pte_t pte) {
	return pte; }
static inline pte_t pte_mkspecial(pte_t pte) {
	return pte; }

/* Atomic PTE updates */
static inline unsigned long pte_update(struct mm_struct *mm,
				       unsigned long addr,
				       pte_t *ptep, unsigned long clr,
				       int huge)
{
	unsigned long old, tmp;

	__asm__ __volatile__(
	"1:	ldarx	%0,0,%3		# pte_update\n\
	andi.	%1,%0,%6\n\
	bne-	1b \n\
	andc	%1,%0,%4 \n\
	stdcx.	%1,0,%3 \n\
	bne-	1b"
	: "=&r" (old), "=&r" (tmp), "=m" (*ptep)
	: "r" (ptep), "r" (clr), "m" (*ptep), "i" (_PAGE_BUSY)
	: "cc" );

	if (old & _PAGE_HASHPTE)
		hpte_need_flush(mm, addr, ptep, old, huge);
	return old;
}

static inline int __ptep_test_and_clear_young(struct mm_struct *mm,
					      unsigned long addr, pte_t *ptep)
{
	unsigned long old;

       	if ((pte_val(*ptep) & (_PAGE_ACCESSED | _PAGE_HASHPTE)) == 0)
		return 0;
	old = pte_update(mm, addr, ptep, _PAGE_ACCESSED, 0);
	return (old & _PAGE_ACCESSED) != 0;
}
#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
#define ptep_test_and_clear_young(__vma, __addr, __ptep)		   \
({									   \
	int __r;							   \
	__r = __ptep_test_and_clear_young((__vma)->vm_mm, __addr, __ptep); \
	__r;								   \
})

#define __HAVE_ARCH_PTEP_SET_WRPROTECT
static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
				      pte_t *ptep)
{
	unsigned long old;

       	if ((pte_val(*ptep) & _PAGE_RW) == 0)
       		return;
	old = pte_update(mm, addr, ptep, _PAGE_RW, 0);
}

/*
 * We currently remove entries from the hashtable regardless of whether
 * the entry was young or dirty. The generic routines only flush if the
 * entry was young or dirty which is not good enough.
 *
 * We should be more intelligent about this but for the moment we override
 * these functions and force a tlb flush unconditionally
 */
#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
#define ptep_clear_flush_young(__vma, __address, __ptep)		\
({									\
	int __young = __ptep_test_and_clear_young((__vma)->vm_mm, __address, \
						  __ptep);		\
	__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)
{
	unsigned long old = pte_update(mm, addr, ptep, ~0UL, 0);
	return __pte(old);
}

static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
			     pte_t * ptep)
{
	pte_update(mm, addr, ptep, ~0UL, 0);
}

/*
 * set_pte stores a linux PTE into the linux page table.
 */
static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
			      pte_t *ptep, pte_t pte)
{
	if (pte_present(*ptep))
		pte_clear(mm, addr, ptep);
	pte = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS);
	*ptep = pte;
}

/* Set the dirty and/or accessed bits atomically in a linux PTE, this
 * function doesn't need to flush the hash entry
 */
#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
static inline void __ptep_set_access_flags(pte_t *ptep, pte_t entry, int dirty)
{
	unsigned long bits = pte_val(entry) &
		(_PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_RW | _PAGE_EXEC);
	unsigned long old, tmp;

	__asm__ __volatile__(
	"1:	ldarx	%0,0,%4\n\
		andi.	%1,%0,%6\n\
		bne-	1b \n\
		or	%0,%3,%0\n\
		stdcx.	%0,0,%4\n\
		bne-	1b"
	:"=&r" (old), "=&r" (tmp), "=m" (*ptep)
	:"r" (bits), "r" (ptep), "m" (*ptep), "i" (_PAGE_BUSY)
	:"cc");
}
#define  ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
({									   \
	int __changed = !pte_same(*(__ptep), __entry);			   \
	if (__changed) {						   \
		__ptep_set_access_flags(__ptep, __entry, __dirty);    	   \
		flush_tlb_page_nohash(__vma, __address);		   \
	}								   \
	__changed;							   \
})

/*
 * Macro to mark a page protection value as "uncacheable".
 */
#define pgprot_noncached(prot)	(__pgprot(pgprot_val(prot) | _PAGE_NO_CACHE | _PAGE_GUARDED))

struct file;
extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
				     unsigned long size, pgprot_t vma_prot);
#define __HAVE_PHYS_MEM_ACCESS_PROT

#define __HAVE_ARCH_PTE_SAME
#define pte_same(A,B)	(((pte_val(A) ^ pte_val(B)) & ~_PAGE_HPTEFLAGS) == 0)

#define pte_ERROR(e) \
	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
#define pmd_ERROR(e) \
	printk("%s:%d: bad pmd %08lx.\n", __FILE__, __LINE__, pmd_val(e))
#define pgd_ERROR(e) \
	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))

/* Encode and de-code a swap entry */
#define __swp_type(entry)	(((entry).val >> 1) & 0x3f)
#define __swp_offset(entry)	((entry).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) >> PTE_RPN_SHIFT})
#define __swp_entry_to_pte(x)	((pte_t) { (x).val << PTE_RPN_SHIFT })
#define pte_to_pgoff(pte)	(pte_val(pte) >> PTE_RPN_SHIFT)
#define pgoff_to_pte(off)	((pte_t) {((off) << PTE_RPN_SHIFT)|_PAGE_FILE})
#define PTE_FILE_MAX_BITS	(BITS_PER_LONG - PTE_RPN_SHIFT)

void pgtable_cache_init(void);

/*
 * find_linux_pte returns the address of a linux pte for a given
 * effective address and directory.  If not found, it returns zero.
 */static inline pte_t *find_linux_pte(pgd_t *pgdir, unsigned long ea)
{
	pgd_t *pg;
	pud_t *pu;
	pmd_t *pm;
	pte_t *pt = NULL;

	pg = pgdir + pgd_index(ea);
	if (!pgd_none(*pg)) {
		pu = pud_offset(pg, ea);
		if (!pud_none(*pu)) {
			pm = pmd_offset(pu, ea);
			if (pmd_present(*pm))
				pt = pte_offset_kernel(pm, ea);
		}
	}
	return pt;
}

#endif /* __ASSEMBLY__ */

#endif /* _ASM_POWERPC_PGTABLE_PPC64_H_ */
Commit	Line	Data
f88df14b DG	1	#ifndef _ASM_POWERPC_PGTABLE_PPC64_H_
	2	#define _ASM_POWERPC_PGTABLE_PPC64_H_
	3	/*
	4	* This file contains the functions and defines necessary to modify and use
	5	* the ppc64 hashed page table.
	6	*/
	7
	8	#ifndef __ASSEMBLY__
	9	#include <linux/stddef.h>
f88df14b	10	#include <asm/tlbflush.h>
f88df14b DG	11	#endif /* __ASSEMBLY__ */
	12
	13	#ifdef CONFIG_PPC_64K_PAGES
	14	#include <asm/pgtable-64k.h>
	15	#else
	16	#include <asm/pgtable-4k.h>
	17	#endif
	18
	19	#define FIRST_USER_ADDRESS 0
	20
	21	/*
	22	* Size of EA range mapped by our pagetables.
	23	*/
	24	#define PGTABLE_EADDR_SIZE (PTE_INDEX_SIZE + PMD_INDEX_SIZE + \
	25	PUD_INDEX_SIZE + PGD_INDEX_SIZE + PAGE_SHIFT)
3d5134ee	26	#define PGTABLE_RANGE (ASM_CONST(1) << PGTABLE_EADDR_SIZE)
f88df14b DG	27
	28	#if TASK_SIZE_USER64 > PGTABLE_RANGE
	29	#error TASK_SIZE_USER64 exceeds pagetable range
	30	#endif
	31
	32	#if TASK_SIZE_USER64 > (1UL << (USER_ESID_BITS + SID_SHIFT))
	33	#error TASK_SIZE_USER64 exceeds user VSID range
	34	#endif
	35
3d5134ee	36
f88df14b DG	37	/*
	38	* Define the address range of the vmalloc VM area.
	39	*/
	40	#define VMALLOC_START ASM_CONST(0xD000000000000000)
3d5134ee	41	#define VMALLOC_SIZE (PGTABLE_RANGE >> 1)
f88df14b DG	42	#define VMALLOC_END (VMALLOC_START + VMALLOC_SIZE)
	43
	44	/*
3d5134ee BH	45	* Define the address ranges for MMIO and IO space :
	46	*
	47	* ISA_IO_BASE = VMALLOC_END, 64K reserved area
	48	* PHB_IO_BASE = ISA_IO_BASE + 64K to ISA_IO_BASE + 2G, PHB IO spaces
	49	* IOREMAP_BASE = ISA_IO_BASE + 2G to VMALLOC_START + PGTABLE_RANGE
f88df14b	50	*/
3d5134ee BH	51	#define FULL_IO_SIZE 0x80000000ul
	52	#define ISA_IO_BASE (VMALLOC_END)
	53	#define ISA_IO_END (VMALLOC_END + 0x10000ul)
	54	#define PHB_IO_BASE (ISA_IO_END)
	55	#define PHB_IO_END (VMALLOC_END + FULL_IO_SIZE)
	56	#define IOREMAP_BASE (PHB_IO_END)
	57	#define IOREMAP_END (VMALLOC_START + PGTABLE_RANGE)
f88df14b DG	58
	59	/*
	60	* Region IDs
	61	*/
	62	#define REGION_SHIFT 60UL
	63	#define REGION_MASK (0xfUL << REGION_SHIFT)
	64	#define REGION_ID(ea) (((unsigned long)(ea)) >> REGION_SHIFT)
	65
	66	#define VMALLOC_REGION_ID (REGION_ID(VMALLOC_START))
	67	#define KERNEL_REGION_ID (REGION_ID(PAGE_OFFSET))
cec08e7a	68	#define VMEMMAP_REGION_ID (0xfUL)
f88df14b DG	69	#define USER_REGION_ID (0UL)
f88df14b DG	70
d29eff7b	71	/*
cec08e7a	72	* Defines the address of the vmemap area, in its own region
d29eff7b	73	*/
cec08e7a BH	74	#define VMEMMAP_BASE (VMEMMAP_REGION_ID << REGION_SHIFT)
	75	#define vmemmap ((struct page *)VMEMMAP_BASE)
	76
d29eff7b	77
f88df14b DG	78	/*
	79	* Common bits in a linux-style PTE. These match the bits in the
	80	* (hardware-defined) PowerPC PTE as closely as possible. Additional
	81	* bits may be defined in pgtable-*.h
	82	*/
	83	#define _PAGE_PRESENT 0x0001 /* software: pte contains a translation */
	84	#define _PAGE_USER 0x0002 /* matches one of the PP bits */
	85	#define _PAGE_FILE 0x0002 /* (!present only) software: pte holds file offset */
	86	#define _PAGE_EXEC 0x0004 /* No execute on POWER4 and newer (we invert) */
	87	#define _PAGE_GUARDED 0x0008
	88	#define _PAGE_COHERENT 0x0010 /* M: enforce memory coherence (SMP systems) */
	89	#define _PAGE_NO_CACHE 0x0020 /* I: cache inhibit */
	90	#define _PAGE_WRITETHRU 0x0040 /* W: cache write-through */
	91	#define _PAGE_DIRTY 0x0080 /* C: page changed */
	92	#define _PAGE_ACCESSED 0x0100 /* R: page referenced */
	93	#define _PAGE_RW 0x0200 /* software: user write access allowed */
	94	#define _PAGE_HASHPTE 0x0400 /* software: pte has an associated HPTE */
	95	#define _PAGE_BUSY 0x0800 /* software: PTE & hash are busy */
	96
	97	#define _PAGE_BASE (_PAGE_PRESENT \| _PAGE_ACCESSED \| _PAGE_COHERENT)
	98
	99	#define _PAGE_WRENABLE (_PAGE_RW \| _PAGE_DIRTY)
	100
	101	/* __pgprot defined in asm-powerpc/page.h */
	102	#define PAGE_NONE __pgprot(_PAGE_PRESENT \| _PAGE_ACCESSED)
	103
	104	#define PAGE_SHARED __pgprot(_PAGE_BASE \| _PAGE_RW \| _PAGE_USER)
	105	#define PAGE_SHARED_X __pgprot(_PAGE_BASE \| _PAGE_RW \| _PAGE_USER \| _PAGE_EXEC)
	106	#define PAGE_COPY __pgprot(_PAGE_BASE \| _PAGE_USER)
	107	#define PAGE_COPY_X __pgprot(_PAGE_BASE \| _PAGE_USER \| _PAGE_EXEC)
	108	#define PAGE_READONLY __pgprot(_PAGE_BASE \| _PAGE_USER)
	109	#define PAGE_READONLY_X __pgprot(_PAGE_BASE \| _PAGE_USER \| _PAGE_EXEC)
	110	#define PAGE_KERNEL __pgprot(_PAGE_BASE \| _PAGE_WRENABLE)
	111	#define PAGE_KERNEL_CI __pgprot(_PAGE_PRESENT \| _PAGE_ACCESSED \| \
	112	_PAGE_WRENABLE \| _PAGE_NO_CACHE \| _PAGE_GUARDED)
	113	#define PAGE_KERNEL_EXEC __pgprot(_PAGE_BASE \| _PAGE_WRENABLE \| _PAGE_EXEC)
	114
	115	#define PAGE_AGP __pgprot(_PAGE_BASE \| _PAGE_WRENABLE \| _PAGE_NO_CACHE)
	116	#define HAVE_PAGE_AGP
	117
	118	/* PTEIDX nibble */
	119	#define _PTEIDX_SECONDARY 0x8
	120	#define _PTEIDX_GROUP_IX 0x7
	121
	122
	123	/*
	124	* POWER4 and newer have per page execute protection, older chips can only
	125	* do this on a segment (256MB) basis.
	126	*
	127	* Also, write permissions imply read permissions.
	128	* This is the closest we can get..
	129	*
	130	* Note due to the way vm flags are laid out, the bits are XWR
	131	*/
	132	#define __P000 PAGE_NONE
	133	#define __P001 PAGE_READONLY
	134	#define __P010 PAGE_COPY
	135	#define __P011 PAGE_COPY
	136	#define __P100 PAGE_READONLY_X
	137	#define __P101 PAGE_READONLY_X
	138	#define __P110 PAGE_COPY_X
	139	#define __P111 PAGE_COPY_X
	140
	141	#define __S000 PAGE_NONE
142	#define __S001 PAGE_READONLY
143	#define __S010 PAGE_SHARED
144	#define __S011 PAGE_SHARED
145	#define __S100 PAGE_READONLY_X
146	#define __S101 PAGE_READONLY_X
147	#define __S110 PAGE_SHARED_X
148	#define __S111 PAGE_SHARED_X
149
f88df14b DG	150	#ifdef CONFIG_HUGETLB_PAGE
	151
	152	#define HAVE_ARCH_UNMAPPED_AREA
	153	#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN
	154
	155	#endif
	156
	157	#ifndef __ASSEMBLY__
	158
	159	/*
	160	* Conversion functions: convert a page and protection to a page entry,
	161	* and a page entry and page directory to the page they refer to.
	162	*
	163	* mk_pte takes a (struct page *) as input
	164	*/
	165	#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
	166
	167	static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
	168	{
	169	pte_t pte;
	170
	171
	172	pte_val(pte) = (pfn << PTE_RPN_SHIFT) \| pgprot_val(pgprot);
	173	return pte;
	174	}
	175
	176	#define pte_modify(_pte, newprot) \
	177	(__pte((pte_val(_pte) & _PAGE_CHG_MASK) \| pgprot_val(newprot)))
	178
	179	#define pte_none(pte) ((pte_val(pte) & ~_PAGE_HPTEFLAGS) == 0)
	180	#define pte_present(pte) (pte_val(pte) & _PAGE_PRESENT)
	181
	182	/* pte_clear moved to later in this file */
	183
	184	#define pte_pfn(x) ((unsigned long)((pte_val(x)>>PTE_RPN_SHIFT)))
	185	#define pte_page(x) pfn_to_page(pte_pfn(x))
	186
	187	#define PMD_BAD_BITS (PTE_TABLE_SIZE-1)
	188	#define PUD_BAD_BITS (PMD_TABLE_SIZE-1)
	189
	190	#define pmd_set(pmdp, pmdval) (pmd_val(*(pmdp)) = (pmdval))
	191	#define pmd_none(pmd) (!pmd_val(pmd))
	192	#define pmd_bad(pmd) (!is_kernel_addr(pmd_val(pmd)) \
	193	\|\| (pmd_val(pmd) & PMD_BAD_BITS))
	194	#define pmd_present(pmd) (pmd_val(pmd) != 0)
	195	#define pmd_clear(pmdp) (pmd_val(*(pmdp)) = 0)
	196	#define pmd_page_vaddr(pmd) (pmd_val(pmd) & ~PMD_MASKED_BITS)
	197	#define pmd_page(pmd) virt_to_page(pmd_page_vaddr(pmd))
	198
	199	#define pud_set(pudp, pudval) (pud_val(*(pudp)) = (pudval))
	200	#define pud_none(pud) (!pud_val(pud))
	201	#define pud_bad(pud) (!is_kernel_addr(pud_val(pud)) \
	202	\|\| (pud_val(pud) & PUD_BAD_BITS))
	203	#define pud_present(pud) (pud_val(pud) != 0)
	204	#define pud_clear(pudp) (pud_val(*(pudp)) = 0)
	205	#define pud_page_vaddr(pud) (pud_val(pud) & ~PUD_MASKED_BITS)
	206	#define pud_page(pud) virt_to_page(pud_page_vaddr(pud))
	207
	208	#define pgd_set(pgdp, pudp) ({pgd_val(*(pgdp)) = (unsigned long)(pudp);})
	209
	210	/*
	211	* Find an entry in a page-table-directory. We combine the address region
	212	* (the high order N bits) and the pgd portion of the address.
	213	*/
214	/* to avoid overflow in free_pgtables we don't use PTRS_PER_PGD here */
215	#define pgd_index(address) (((address) >> (PGDIR_SHIFT)) & 0x1ff)
216
217	#define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address))
218
219	#define pmd_offset(pudp,addr) \
220	(((pmd_t ) pud_page_vaddr((pudp))) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
221
222	#define pte_offset_kernel(dir,addr) \
223	(((pte_t ) pmd_page_vaddr((dir))) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)))
224
225	#define pte_offset_map(dir,addr) pte_offset_kernel((dir), (addr))
226	#define pte_offset_map_nested(dir,addr) pte_offset_kernel((dir), (addr))
227	#define pte_unmap(pte) do { } while(0)
228	#define pte_unmap_nested(pte) do { } while(0)
229
230	/* to find an entry in a kernel page-table-directory */
231	/* This now only contains the vmalloc pages */
232	#define pgd_offset_k(address) pgd_offset(&init_mm, address)
233
234	/*
235	* The following only work if pte_present() is true.
236	* Undefined behaviour if not..
237	*/
f88df14b	238	static inline int pte_write(pte_t pte) { return pte_val(pte) & _PAGE_RW;}
f88df14b DG	239	static inline int pte_dirty(pte_t pte) { return pte_val(pte) & _PAGE_DIRTY;}
	240	static inline int pte_young(pte_t pte) { return pte_val(pte) & _PAGE_ACCESSED;}
	241	static inline int pte_file(pte_t pte) { return pte_val(pte) & _PAGE_FILE;}
7e675137	242	static inline int pte_special(pte_t pte) { return 0; }
f88df14b DG	243
	244	static inline void pte_uncache(pte_t pte) { pte_val(pte) \|= _PAGE_NO_CACHE; }
	245	static inline void pte_cache(pte_t pte) { pte_val(pte) &= ~_PAGE_NO_CACHE; }
	246
f88df14b DG	247	static inline pte_t pte_wrprotect(pte_t pte) {
	248	pte_val(pte) &= ~(_PAGE_RW); return pte; }
	249	static inline pte_t pte_mkclean(pte_t pte) {
	250	pte_val(pte) &= ~(_PAGE_DIRTY); return pte; }
	251	static inline pte_t pte_mkold(pte_t pte) {
	252	pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
f88df14b DG	253	static inline pte_t pte_mkwrite(pte_t pte) {
	254	pte_val(pte) \|= _PAGE_RW; return pte; }
	255	static inline pte_t pte_mkdirty(pte_t pte) {
	256	pte_val(pte) \|= _PAGE_DIRTY; return pte; }
	257	static inline pte_t pte_mkyoung(pte_t pte) {
	258	pte_val(pte) \|= _PAGE_ACCESSED; return pte; }
	259	static inline pte_t pte_mkhuge(pte_t pte) {
	260	return pte; }
7e675137 NP	261	static inline pte_t pte_mkspecial(pte_t pte) {
7e675137 NP	262	return pte; }
f88df14b DG	263
	264	/* Atomic PTE updates */
	265	static inline unsigned long pte_update(struct mm_struct *mm,
	266	unsigned long addr,
	267	pte_t *ptep, unsigned long clr,
	268	int huge)
	269	{
	270	unsigned long old, tmp;
	271
	272	__asm__ __volatile__(
	273	"1: ldarx %0,0,%3 # pte_update\n\
	274	andi. %1,%0,%6\n\
	275	bne- 1b \n\
	276	andc %1,%0,%4 \n\
	277	stdcx. %1,0,%3 \n\
	278	bne- 1b"
	279	: "=&r" (old), "=&r" (tmp), "=m" (*ptep)
	280	: "r" (ptep), "r" (clr), "m" (*ptep), "i" (_PAGE_BUSY)
	281	: "cc" );
	282
	283	if (old & _PAGE_HASHPTE)
	284	hpte_need_flush(mm, addr, ptep, old, huge);
	285	return old;
	286	}
	287
	288	static inline int __ptep_test_and_clear_young(struct mm_struct *mm,
	289	unsigned long addr, pte_t *ptep)
	290	{
	291	unsigned long old;
	292
	293	if ((pte_val(*ptep) & (_PAGE_ACCESSED \| _PAGE_HASHPTE)) == 0)
	294	return 0;
	295	old = pte_update(mm, addr, ptep, _PAGE_ACCESSED, 0);
	296	return (old & _PAGE_ACCESSED) != 0;
	297	}
	298	#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
	299	#define ptep_test_and_clear_young(__vma, __addr, __ptep) \
	300	({ \
	301	int __r; \
	302	__r = __ptep_test_and_clear_young((__vma)->vm_mm, __addr, __ptep); \
	303	__r; \
	304	})
	305
f88df14b DG	306	#define __HAVE_ARCH_PTEP_SET_WRPROTECT
	307	static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
	308	pte_t *ptep)
	309	{
	310	unsigned long old;
	311
	312	if ((pte_val(*ptep) & _PAGE_RW) == 0)
	313	return;
	314	old = pte_update(mm, addr, ptep, _PAGE_RW, 0);
	315	}
	316
	317	/*
	318	* We currently remove entries from the hashtable regardless of whether
	319	* the entry was young or dirty. The generic routines only flush if the
	320	* entry was young or dirty which is not good enough.
	321	*
	322	* We should be more intelligent about this but for the moment we override
	323	* these functions and force a tlb flush unconditionally
	324	*/
	325	#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
	326	#define ptep_clear_flush_young(__vma, __address, __ptep) \
	327	({ \
	328	int __young = __ptep_test_and_clear_young((__vma)->vm_mm, __address, \
	329	__ptep); \
	330	__young; \
	331	})
	332
f88df14b DG	333	#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
	334	static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
	335	unsigned long addr, pte_t *ptep)
	336	{
	337	unsigned long old = pte_update(mm, addr, ptep, ~0UL, 0);
	338	return __pte(old);
	339	}
	340
	341	static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
	342	pte_t * ptep)
	343	{
	344	pte_update(mm, addr, ptep, ~0UL, 0);
	345	}
	346
	347	/*
	348	* set_pte stores a linux PTE into the linux page table.
	349	*/
	350	static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
	351	pte_t *ptep, pte_t pte)
	352	{
	353	if (pte_present(*ptep))
	354	pte_clear(mm, addr, ptep);
	355	pte = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS);
	356	*ptep = pte;
	357	}
	358
	359	/* Set the dirty and/or accessed bits atomically in a linux PTE, this
	360	* function doesn't need to flush the hash entry
	361	*/
	362	#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
	363	static inline void __ptep_set_access_flags(pte_t *ptep, pte_t entry, int dirty)
	364	{
	365	unsigned long bits = pte_val(entry) &
	366	(_PAGE_DIRTY \| _PAGE_ACCESSED \| _PAGE_RW \| _PAGE_EXEC);
	367	unsigned long old, tmp;
	368
	369	__asm__ __volatile__(
	370	"1: ldarx %0,0,%4\n\
	371	andi. %1,%0,%6\n\
	372	bne- 1b \n\
	373	or %0,%3,%0\n\
	374	stdcx. %0,0,%4\n\
	375	bne- 1b"
	376	:"=&r" (old), "=&r" (tmp), "=m" (*ptep)
	377	:"r" (bits), "r" (ptep), "m" (*ptep), "i" (_PAGE_BUSY)
	378	:"cc");
	379	}
	380	#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
8dab5241 BH	381	({ \
	382	int __changed = !pte_same(*(__ptep), __entry); \
	383	if (__changed) { \
	384	__ptep_set_access_flags(__ptep, __entry, __dirty); \
	385	flush_tlb_page_nohash(__vma, __address); \
	386	} \
	387	__changed; \
	388	})
f88df14b DG	389
	390	/*
	391	* Macro to mark a page protection value as "uncacheable".
	392	*/
	393	#define pgprot_noncached(prot) (__pgprot(pgprot_val(prot) \| _PAGE_NO_CACHE \| _PAGE_GUARDED))
	394
	395	struct file;
	396	extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
	397	unsigned long size, pgprot_t vma_prot);
	398	#define __HAVE_PHYS_MEM_ACCESS_PROT
	399
	400	#define __HAVE_ARCH_PTE_SAME
	401	#define pte_same(A,B) (((pte_val(A) ^ pte_val(B)) & ~_PAGE_HPTEFLAGS) == 0)
	402
	403	#define pte_ERROR(e) \
	404	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
	405	#define pmd_ERROR(e) \
	406	printk("%s:%d: bad pmd %08lx.\n", __FILE__, __LINE__, pmd_val(e))
	407	#define pgd_ERROR(e) \
	408	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
	409
f88df14b DG	410	/* Encode and de-code a swap entry */
	411	#define __swp_type(entry) (((entry).val >> 1) & 0x3f)
	412	#define __swp_offset(entry) ((entry).val >> 8)
	413	#define __swp_entry(type, offset) ((swp_entry_t){((type)<< 1)\|((offset)<<8)})
	414	#define __pte_to_swp_entry(pte) ((swp_entry_t){pte_val(pte) >> PTE_RPN_SHIFT})
	415	#define __swp_entry_to_pte(x) ((pte_t) { (x).val << PTE_RPN_SHIFT })
	416	#define pte_to_pgoff(pte) (pte_val(pte) >> PTE_RPN_SHIFT)
	417	#define pgoff_to_pte(off) ((pte_t) {((off) << PTE_RPN_SHIFT)\|_PAGE_FILE})
	418	#define PTE_FILE_MAX_BITS (BITS_PER_LONG - PTE_RPN_SHIFT)
	419
f88df14b DG	420	void pgtable_cache_init(void);
	421
	422	/*
	423	* find_linux_pte returns the address of a linux pte for a given
	424	* effective address and directory. If not found, it returns zero.
	425	/static inline pte_t find_linux_pte(pgd_t *pgdir, unsigned long ea)
	426	{
	427	pgd_t *pg;
	428	pud_t *pu;
	429	pmd_t *pm;
	430	pte_t *pt = NULL;
	431
	432	pg = pgdir + pgd_index(ea);
	433	if (!pgd_none(*pg)) {
	434	pu = pud_offset(pg, ea);
	435	if (!pud_none(*pu)) {
	436	pm = pmd_offset(pu, ea);
	437	if (pmd_present(*pm))
	438	pt = pte_offset_kernel(pm, ea);
	439	}
	440	}
	441	return pt;
	442	}
	443
	444	#endif /* __ASSEMBLY__ */
	445
	446	#endif /* _ASM_POWERPC_PGTABLE_PPC64_H_ */