[net-next-2.6.git] / mm / pagewalk.c

#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/sched.h>
#include <linux/hugetlb.h>

static int walk_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
			  struct mm_walk *walk)
{
	pte_t *pte;
	int err = 0;

	pte = pte_offset_map(pmd, addr);
	for (;;) {
		err = walk->pte_entry(pte, addr, addr + PAGE_SIZE, walk);
		if (err)
		       break;
		addr += PAGE_SIZE;
		if (addr == end)
			break;
		pte++;
	}

	pte_unmap(pte);
	return err;
}

static int walk_pmd_range(pud_t *pud, unsigned long addr, unsigned long end,
			  struct mm_walk *walk)
{
	pmd_t *pmd;
	unsigned long next;
	int err = 0;

	pmd = pmd_offset(pud, addr);
	do {
		next = pmd_addr_end(addr, end);
		if (pmd_none_or_clear_bad(pmd)) {
			if (walk->pte_hole)
				err = walk->pte_hole(addr, next, walk);
			if (err)
				break;
			continue;
		}
		if (walk->pmd_entry)
			err = walk->pmd_entry(pmd, addr, next, walk);
		if (!err && walk->pte_entry)
			err = walk_pte_range(pmd, addr, next, walk);
		if (err)
			break;
	} while (pmd++, addr = next, addr != end);

	return err;
}

static int walk_pud_range(pgd_t *pgd, unsigned long addr, unsigned long end,
			  struct mm_walk *walk)
{
	pud_t *pud;
	unsigned long next;
	int err = 0;

	pud = pud_offset(pgd, addr);
	do {
		next = pud_addr_end(addr, end);
		if (pud_none_or_clear_bad(pud)) {
			if (walk->pte_hole)
				err = walk->pte_hole(addr, next, walk);
			if (err)
				break;
			continue;
		}
		if (walk->pud_entry)
			err = walk->pud_entry(pud, addr, next, walk);
		if (!err && (walk->pmd_entry || walk->pte_entry))
			err = walk_pmd_range(pud, addr, next, walk);
		if (err)
			break;
	} while (pud++, addr = next, addr != end);

	return err;
}

#ifdef CONFIG_HUGETLB_PAGE
static unsigned long hugetlb_entry_end(struct hstate *h, unsigned long addr,
				       unsigned long end)
{
	unsigned long boundary = (addr & huge_page_mask(h)) + huge_page_size(h);
	return boundary < end ? boundary : end;
}

static int walk_hugetlb_range(struct vm_area_struct *vma,
			      unsigned long addr, unsigned long end,
			      struct mm_walk *walk)
{
	struct hstate *h = hstate_vma(vma);
	unsigned long next;
	unsigned long hmask = huge_page_mask(h);
	pte_t *pte;
	int err = 0;

	do {
		next = hugetlb_entry_end(h, addr, end);
		pte = huge_pte_offset(walk->mm, addr & hmask);
		if (pte && walk->hugetlb_entry)
			err = walk->hugetlb_entry(pte, hmask, addr, next, walk);
		if (err)
			return err;
	} while (addr = next, addr != end);

	return 0;
}
#endif

/**
 * walk_page_range - walk a memory map's page tables with a callback
 * @mm: memory map to walk
 * @addr: starting address
 * @end: ending address
 * @walk: set of callbacks to invoke for each level of the tree
 *
 * Recursively walk the page table for the memory area in a VMA,
 * calling supplied callbacks. Callbacks are called in-order (first
 * PGD, first PUD, first PMD, first PTE, second PTE... second PMD,
 * etc.). If lower-level callbacks are omitted, walking depth is reduced.
 *
 * Each callback receives an entry pointer and the start and end of the
 * associated range, and a copy of the original mm_walk for access to
 * the ->private or ->mm fields.
 *
 * No locks are taken, but the bottom level iterator will map PTE
 * directories from highmem if necessary.
 *
 * If any callback returns a non-zero value, the walk is aborted and
 * the return value is propagated back to the caller. Otherwise 0 is returned.
 */
int walk_page_range(unsigned long addr, unsigned long end,
		    struct mm_walk *walk)
{
	pgd_t *pgd;
	unsigned long next;
	int err = 0;
	struct vm_area_struct *vma;

	if (addr >= end)
		return err;

	if (!walk->mm)
		return -EINVAL;

	pgd = pgd_offset(walk->mm, addr);
	do {
		next = pgd_addr_end(addr, end);

		/*
		 * handle hugetlb vma individually because pagetable walk for
		 * the hugetlb page is dependent on the architecture and
		 * we can't handled it in the same manner as non-huge pages.
		 */
		vma = find_vma(walk->mm, addr);
#ifdef CONFIG_HUGETLB_PAGE
		if (vma && is_vm_hugetlb_page(vma)) {
			if (vma->vm_end < next)
				next = vma->vm_end;
			/*
			 * Hugepage is very tightly coupled with vma, so
			 * walk through hugetlb entries within a given vma.
			 */
			err = walk_hugetlb_range(vma, addr, next, walk);
			if (err)
				break;
			pgd = pgd_offset(walk->mm, next);
			continue;
		}
#endif
		if (pgd_none_or_clear_bad(pgd)) {
			if (walk->pte_hole)
				err = walk->pte_hole(addr, next, walk);
			if (err)
				break;
			pgd++;
			continue;
		}
		if (walk->pgd_entry)
			err = walk->pgd_entry(pgd, addr, next, walk);
		if (!err &&
		    (walk->pud_entry || walk->pmd_entry || walk->pte_entry))
			err = walk_pud_range(pgd, addr, next, walk);
		if (err)
			break;
		pgd++;
	} while (addr = next, addr != end);

	return err;
}
Commit	Line	Data
e6473092 MM	1	#include <linux/mm.h>
	2	#include <linux/highmem.h>
	3	#include <linux/sched.h>
d33b9f45	4	#include <linux/hugetlb.h>
e6473092 MM	5
e6473092 MM	6	static int walk_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
2165009b	7	struct mm_walk *walk)
e6473092 MM	8	{
	9	pte_t *pte;
	10	int err = 0;
	11
	12	pte = pte_offset_map(pmd, addr);
556637cd	13	for (;;) {
2165009b	14	err = walk->pte_entry(pte, addr, addr + PAGE_SIZE, walk);
e6473092 MM	15	if (err)
e6473092 MM	16	break;
556637cd JW	17	addr += PAGE_SIZE;
	18	if (addr == end)
	19	break;
	20	pte++;
	21	}
e6473092 MM	22
	23	pte_unmap(pte);
	24	return err;
	25	}
	26
	27	static int walk_pmd_range(pud_t *pud, unsigned long addr, unsigned long end,
2165009b	28	struct mm_walk *walk)
e6473092 MM	29	{
	30	pmd_t *pmd;
	31	unsigned long next;
	32	int err = 0;
	33
	34	pmd = pmd_offset(pud, addr);
	35	do {
	36	next = pmd_addr_end(addr, end);
	37	if (pmd_none_or_clear_bad(pmd)) {
	38	if (walk->pte_hole)
2165009b	39	err = walk->pte_hole(addr, next, walk);
e6473092 MM	40	if (err)
	41	break;
	42	continue;
	43	}
	44	if (walk->pmd_entry)
2165009b	45	err = walk->pmd_entry(pmd, addr, next, walk);
e6473092	46	if (!err && walk->pte_entry)
2165009b	47	err = walk_pte_range(pmd, addr, next, walk);
e6473092 MM	48	if (err)
	49	break;
	50	} while (pmd++, addr = next, addr != end);
	51
	52	return err;
	53	}
	54
	55	static int walk_pud_range(pgd_t *pgd, unsigned long addr, unsigned long end,
2165009b	56	struct mm_walk *walk)
e6473092 MM	57	{
	58	pud_t *pud;
	59	unsigned long next;
	60	int err = 0;
	61
	62	pud = pud_offset(pgd, addr);
	63	do {
	64	next = pud_addr_end(addr, end);
	65	if (pud_none_or_clear_bad(pud)) {
	66	if (walk->pte_hole)
2165009b	67	err = walk->pte_hole(addr, next, walk);
e6473092 MM	68	if (err)
	69	break;
	70	continue;
	71	}
	72	if (walk->pud_entry)
2165009b	73	err = walk->pud_entry(pud, addr, next, walk);
e6473092	74	if (!err && (walk->pmd_entry \|\| walk->pte_entry))
2165009b	75	err = walk_pmd_range(pud, addr, next, walk);
e6473092 MM	76	if (err)
	77	break;
	78	} while (pud++, addr = next, addr != end);
	79
	80	return err;
	81	}
	82
116354d1 NH	83	#ifdef CONFIG_HUGETLB_PAGE
	84	static unsigned long hugetlb_entry_end(struct hstate *h, unsigned long addr,
	85	unsigned long end)
	86	{
	87	unsigned long boundary = (addr & huge_page_mask(h)) + huge_page_size(h);
	88	return boundary < end ? boundary : end;
	89	}
	90
	91	static int walk_hugetlb_range(struct vm_area_struct *vma,
	92	unsigned long addr, unsigned long end,
	93	struct mm_walk *walk)
	94	{
	95	struct hstate *h = hstate_vma(vma);
	96	unsigned long next;
	97	unsigned long hmask = huge_page_mask(h);
	98	pte_t *pte;
	99	int err = 0;
	100
	101	do {
	102	next = hugetlb_entry_end(h, addr, end);
	103	pte = huge_pte_offset(walk->mm, addr & hmask);
	104	if (pte && walk->hugetlb_entry)
	105	err = walk->hugetlb_entry(pte, hmask, addr, next, walk);
	106	if (err)
	107	return err;
	108	} while (addr = next, addr != end);
	109
	110	return 0;
	111	}
	112	#endif
	113
e6473092 MM	114	/**
e6473092 MM	115	* walk_page_range - walk a memory map's page tables with a callback
7682486b RD	116	* @mm: memory map to walk
	117	* @addr: starting address
	118	* @end: ending address
	119	* @walk: set of callbacks to invoke for each level of the tree
e6473092 MM	120	*
	121	* Recursively walk the page table for the memory area in a VMA,
	122	* calling supplied callbacks. Callbacks are called in-order (first
	123	* PGD, first PUD, first PMD, first PTE, second PTE... second PMD,
	124	* etc.). If lower-level callbacks are omitted, walking depth is reduced.
	125	*
2165009b DH	126	* Each callback receives an entry pointer and the start and end of the
	127	* associated range, and a copy of the original mm_walk for access to
	128	* the ->private or ->mm fields.
e6473092 MM	129	*
	130	* No locks are taken, but the bottom level iterator will map PTE
	131	* directories from highmem if necessary.
	132	*
	133	* If any callback returns a non-zero value, the walk is aborted and
	134	* the return value is propagated back to the caller. Otherwise 0 is returned.
	135	*/
2165009b DH	136	int walk_page_range(unsigned long addr, unsigned long end,
2165009b DH	137	struct mm_walk *walk)
e6473092 MM	138	{
	139	pgd_t *pgd;
	140	unsigned long next;
	141	int err = 0;
d33b9f45	142	struct vm_area_struct *vma;
e6473092 MM	143
	144	if (addr >= end)
	145	return err;
	146
2165009b DH	147	if (!walk->mm)
	148	return -EINVAL;
	149
	150	pgd = pgd_offset(walk->mm, addr);
e6473092 MM	151	do {
e6473092 MM	152	next = pgd_addr_end(addr, end);
d33b9f45	153
5dc37642 NH	154	/*
	155	* handle hugetlb vma individually because pagetable walk for
	156	* the hugetlb page is dependent on the architecture and
	157	* we can't handled it in the same manner as non-huge pages.
	158	*/
d33b9f45	159	vma = find_vma(walk->mm, addr);
5dc37642	160	#ifdef CONFIG_HUGETLB_PAGE
d33b9f45 NH	161	if (vma && is_vm_hugetlb_page(vma)) {
	162	if (vma->vm_end < next)
	163	next = vma->vm_end;
116354d1 NH	164	/*
	165	* Hugepage is very tightly coupled with vma, so
	166	* walk through hugetlb entries within a given vma.
	167	*/
	168	err = walk_hugetlb_range(vma, addr, next, walk);
5dc37642 NH	169	if (err)
5dc37642 NH	170	break;
116354d1	171	pgd = pgd_offset(walk->mm, next);
d33b9f45 NH	172	continue;
d33b9f45 NH	173	}
5dc37642	174	#endif
e6473092 MM	175	if (pgd_none_or_clear_bad(pgd)) {
e6473092 MM	176	if (walk->pte_hole)
2165009b	177	err = walk->pte_hole(addr, next, walk);
e6473092 MM	178	if (err)
e6473092 MM	179	break;
d33b9f45	180	pgd++;
e6473092 MM	181	continue;
	182	}
	183	if (walk->pgd_entry)
2165009b	184	err = walk->pgd_entry(pgd, addr, next, walk);
e6473092 MM	185	if (!err &&
e6473092 MM	186	(walk->pud_entry \|\| walk->pmd_entry \|\| walk->pte_entry))
2165009b	187	err = walk_pud_range(pgd, addr, next, walk);
e6473092 MM	188	if (err)
e6473092 MM	189	break;
d33b9f45 NH	190	pgd++;
d33b9f45 NH	191	} while (addr = next, addr != end);
e6473092 MM	192
	193	return err;
	194	}