[mirror_ubuntu-bionic-kernel.git] / arch / powerpc / mm / dump_linuxpagetables.c

/*
 * Copyright 2016, Rashmica Gupta, IBM Corp.
 *
 * This traverses the kernel pagetables and dumps the
 * information about the used sections of memory to
 * /sys/kernel/debug/kernel_pagetables.
 *
 * Derived from the arm64 implementation:
 * Copyright (c) 2014, The Linux Foundation, Laura Abbott.
 * (C) Copyright 2008 Intel Corporation, Arjan van de Ven.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; version 2
 * of the License.
 */
#include <linux/debugfs.h>
#include <linux/fs.h>
#include <linux/io.h>
#include <linux/mm.h>
#include <linux/sched.h>
#include <linux/seq_file.h>
#include <asm/fixmap.h>
#include <asm/pgtable.h>
#include <linux/const.h>
#include <asm/page.h>
#include <asm/pgalloc.h>

/*
 * To visualise what is happening,
 *
 *  - PTRS_PER_P** = how many entries there are in the corresponding P**
 *  - P**_SHIFT = how many bits of the address we use to index into the
 * corresponding P**
 *  - P**_SIZE is how much memory we can access through the table - not the
 * size of the table itself.
 * P**={PGD, PUD, PMD, PTE}
 *
 *
 * Each entry of the PGD points to a PUD. Each entry of a PUD points to a
 * PMD. Each entry of a PMD points to a PTE. And every PTE entry points to
 * a page.
 *
 * In the case where there are only 3 levels, the PUD is folded into the
 * PGD: every PUD has only one entry which points to the PMD.
 *
 * The page dumper groups page table entries of the same type into a single
 * description. It uses pg_state to track the range information while
 * iterating over the PTE entries. When the continuity is broken it then
 * dumps out a description of the range - ie PTEs that are virtually contiguous
 * with the same PTE flags are chunked together. This is to make it clear how
 * different areas of the kernel virtual memory are used.
 *
 */
struct pg_state {
	struct seq_file *seq;
	const struct addr_marker *marker;
	unsigned long start_address;
	unsigned int level;
	u64 current_flags;
};

struct addr_marker {
	unsigned long start_address;
	const char *name;
};

static struct addr_marker address_markers[] = {
	{ 0,	"Start of kernel VM" },
	{ 0,	"vmalloc() Area" },
	{ 0,	"vmalloc() End" },
	{ 0,	"isa I/O start" },
	{ 0,	"isa I/O end" },
	{ 0,	"phb I/O start" },
	{ 0,	"phb I/O end" },
	{ 0,	"I/O remap start" },
	{ 0,	"I/O remap end" },
	{ 0,	"vmemmap start" },
	{ -1,	NULL },
};

struct flag_info {
	u64		mask;
	u64		val;
	const char	*set;
	const char	*clear;
	bool		is_val;
	int		shift;
};

static const struct flag_info flag_array[] = {
	{
#ifdef CONFIG_PPC_STD_MMU_64
		.mask	= _PAGE_PRIVILEGED,
		.val	= 0,
#else
		.mask	= _PAGE_USER,
		.val	= _PAGE_USER,
#endif
		.set	= "user",
		.clear	= "    ",
	}, {
		.mask	= _PAGE_RW,
		.val	= _PAGE_RW,
		.set	= "rw",
		.clear	= "ro",
	}, {
		.mask	= _PAGE_EXEC,
		.val	= _PAGE_EXEC,
		.set	= " X ",
		.clear	= "   ",
	}, {
		.mask	= _PAGE_PTE,
		.val	= _PAGE_PTE,
		.set	= "pte",
		.clear	= "   ",
	}, {
		.mask	= _PAGE_PRESENT,
		.val	= _PAGE_PRESENT,
		.set	= "present",
		.clear	= "       ",
	}, {
#ifdef CONFIG_PPC_STD_MMU_64
		.mask	= H_PAGE_HASHPTE,
		.val	= H_PAGE_HASHPTE,
#else
		.mask	= _PAGE_HASHPTE,
		.val	= _PAGE_HASHPTE,
#endif
		.set	= "hpte",
		.clear	= "    ",
	}, {
#ifndef CONFIG_PPC_STD_MMU_64
		.mask	= _PAGE_GUARDED,
		.val	= _PAGE_GUARDED,
		.set	= "guarded",
		.clear	= "       ",
	}, {
#endif
		.mask	= _PAGE_DIRTY,
		.val	= _PAGE_DIRTY,
		.set	= "dirty",
		.clear	= "     ",
	}, {
		.mask	= _PAGE_ACCESSED,
		.val	= _PAGE_ACCESSED,
		.set	= "accessed",
		.clear	= "        ",
	}, {
#ifndef CONFIG_PPC_STD_MMU_64
		.mask	= _PAGE_WRITETHRU,
		.val	= _PAGE_WRITETHRU,
		.set	= "write through",
		.clear	= "             ",
	}, {
#endif
		.mask	= _PAGE_NO_CACHE,
		.val	= _PAGE_NO_CACHE,
		.set	= "no cache",
		.clear	= "        ",
	}, {
#ifdef CONFIG_PPC_BOOK3S_64
		.mask	= H_PAGE_BUSY,
		.val	= H_PAGE_BUSY,
		.set	= "busy",
	}, {
#ifdef CONFIG_PPC_64K_PAGES
		.mask	= H_PAGE_COMBO,
		.val	= H_PAGE_COMBO,
		.set	= "combo",
	}, {
		.mask	= H_PAGE_4K_PFN,
		.val	= H_PAGE_4K_PFN,
		.set	= "4K_pfn",
	}, {
#endif
		.mask	= H_PAGE_F_GIX,
		.val	= H_PAGE_F_GIX,
		.set	= "f_gix",
		.is_val	= true,
		.shift	= H_PAGE_F_GIX_SHIFT,
	}, {
		.mask	= H_PAGE_F_SECOND,
		.val	= H_PAGE_F_SECOND,
		.set	= "f_second",
	}, {
#endif
		.mask	= _PAGE_SPECIAL,
		.val	= _PAGE_SPECIAL,
		.set	= "special",
	}
};

struct pgtable_level {
	const struct flag_info *flag;
	size_t num;
	u64 mask;
};

static struct pgtable_level pg_level[] = {
	{
	}, { /* pgd */
		.flag	= flag_array,
		.num	= ARRAY_SIZE(flag_array),
	}, { /* pud */
		.flag	= flag_array,
		.num	= ARRAY_SIZE(flag_array),
	}, { /* pmd */
		.flag	= flag_array,
		.num	= ARRAY_SIZE(flag_array),
	}, { /* pte */
		.flag	= flag_array,
		.num	= ARRAY_SIZE(flag_array),
	},
};

static void dump_flag_info(struct pg_state *st, const struct flag_info
		*flag, u64 pte, int num)
{
	unsigned int i;

	for (i = 0; i < num; i++, flag++) {
		const char *s = NULL;
		u64 val;

		/* flag not defined so don't check it */
		if (flag->mask == 0)
			continue;
		/* Some 'flags' are actually values */
		if (flag->is_val) {
			val = pte & flag->val;
			if (flag->shift)
				val = val >> flag->shift;
			seq_printf(st->seq, "  %s:%llx", flag->set, val);
		} else {
			if ((pte & flag->mask) == flag->val)
				s = flag->set;
			else
				s = flag->clear;
			if (s)
				seq_printf(st->seq, "  %s", s);
		}
		st->current_flags &= ~flag->mask;
	}
	if (st->current_flags != 0)
		seq_printf(st->seq, "  unknown flags:%llx", st->current_flags);
}

static void dump_addr(struct pg_state *st, unsigned long addr)
{
	static const char units[] = "KMGTPE";
	const char *unit = units;
	unsigned long delta;

	seq_printf(st->seq, "0x%016lx-0x%016lx   ", st->start_address, addr-1);
	delta = (addr - st->start_address) >> 10;
	/* Work out what appropriate unit to use */
	while (!(delta & 1023) && unit[1]) {
		delta >>= 10;
		unit++;
	}
	seq_printf(st->seq, "%9lu%c", delta, *unit);

}

static void note_page(struct pg_state *st, unsigned long addr,
	       unsigned int level, u64 val)
{
	u64 flag = val & pg_level[level].mask;
	/* At first no level is set */
	if (!st->level) {
		st->level = level;
		st->current_flags = flag;
		st->start_address = addr;
		seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
	/*
	 * Dump the section of virtual memory when:
	 *   - the PTE flags from one entry to the next differs.
	 *   - we change levels in the tree.
	 *   - the address is in a different section of memory and is thus
	 *   used for a different purpose, regardless of the flags.
	 */
	} else if (flag != st->current_flags || level != st->level ||
		   addr >= st->marker[1].start_address) {

		/* Check the PTE flags */
		if (st->current_flags) {
			dump_addr(st, addr);

			/* Dump all the flags */
			if (pg_level[st->level].flag)
				dump_flag_info(st, pg_level[st->level].flag,
					  st->current_flags,
					  pg_level[st->level].num);

			seq_puts(st->seq, "\n");
		}

		/*
		 * Address indicates we have passed the end of the
		 * current section of virtual memory
		 */
		while (addr >= st->marker[1].start_address) {
			st->marker++;
			seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
		}
		st->start_address = addr;
		st->current_flags = flag;
		st->level = level;
	}
}

static void walk_pte(struct pg_state *st, pmd_t *pmd, unsigned long start)
{
	pte_t *pte = pte_offset_kernel(pmd, 0);
	unsigned long addr;
	unsigned int i;

	for (i = 0; i < PTRS_PER_PTE; i++, pte++) {
		addr = start + i * PAGE_SIZE;
		note_page(st, addr, 4, pte_val(*pte));

	}
}

static void walk_pmd(struct pg_state *st, pud_t *pud, unsigned long start)
{
	pmd_t *pmd = pmd_offset(pud, 0);
	unsigned long addr;
	unsigned int i;

	for (i = 0; i < PTRS_PER_PMD; i++, pmd++) {
		addr = start + i * PMD_SIZE;
		if (!pmd_none(*pmd))
			/* pmd exists */
			walk_pte(st, pmd, addr);
		else
			note_page(st, addr, 3, pmd_val(*pmd));
	}
}

static void walk_pud(struct pg_state *st, pgd_t *pgd, unsigned long start)
{
	pud_t *pud = pud_offset(pgd, 0);
	unsigned long addr;
	unsigned int i;

	for (i = 0; i < PTRS_PER_PUD; i++, pud++) {
		addr = start + i * PUD_SIZE;
		if (!pud_none(*pud))
			/* pud exists */
			walk_pmd(st, pud, addr);
		else
			note_page(st, addr, 2, pud_val(*pud));
	}
}

static void walk_pagetables(struct pg_state *st)
{
	pgd_t *pgd = pgd_offset_k(0UL);
	unsigned int i;
	unsigned long addr;

	/*
	 * Traverse the linux pagetable structure and dump pages that are in
	 * the hash pagetable.
	 */
	for (i = 0; i < PTRS_PER_PGD; i++, pgd++) {
		addr = KERN_VIRT_START + i * PGDIR_SIZE;
		if (!pgd_none(*pgd))
			/* pgd exists */
			walk_pud(st, pgd, addr);
		else
			note_page(st, addr, 1, pgd_val(*pgd));
	}
}

static void populate_markers(void)
{
	address_markers[0].start_address = PAGE_OFFSET;
	address_markers[1].start_address = VMALLOC_START;
	address_markers[2].start_address = VMALLOC_END;
	address_markers[3].start_address = ISA_IO_BASE;
	address_markers[4].start_address = ISA_IO_END;
	address_markers[5].start_address = PHB_IO_BASE;
	address_markers[6].start_address = PHB_IO_END;
	address_markers[7].start_address = IOREMAP_BASE;
	address_markers[8].start_address = IOREMAP_END;
#ifdef CONFIG_PPC_STD_MMU_64
	address_markers[9].start_address =  H_VMEMMAP_BASE;
#else
	address_markers[9].start_address =  VMEMMAP_BASE;
#endif
}

static int ptdump_show(struct seq_file *m, void *v)
{
	struct pg_state st = {
		.seq = m,
		.start_address = KERN_VIRT_START,
		.marker = address_markers,
	};
	/* Traverse kernel page tables */
	walk_pagetables(&st);
	note_page(&st, 0, 0, 0);
	return 0;
}


static int ptdump_open(struct inode *inode, struct file *file)
{
	return single_open(file, ptdump_show, NULL);
}

static const struct file_operations ptdump_fops = {
	.open		= ptdump_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= single_release,
};

static void build_pgtable_complete_mask(void)
{
	unsigned int i, j;

	for (i = 0; i < ARRAY_SIZE(pg_level); i++)
		if (pg_level[i].flag)
			for (j = 0; j < pg_level[i].num; j++)
				pg_level[i].mask |= pg_level[i].flag[j].mask;
}

static int ptdump_init(void)
{
	struct dentry *debugfs_file;

	populate_markers();
	build_pgtable_complete_mask();
	debugfs_file = debugfs_create_file("kernel_pagetables", 0400, NULL,
			NULL, &ptdump_fops);
	return debugfs_file ? 0 : -ENOMEM;
}
device_initcall(ptdump_init);
Commit	Line	Data
8eb07b18 RG	1	/*
	2	* Copyright 2016, Rashmica Gupta, IBM Corp.
	3	*
	4	* This traverses the kernel pagetables and dumps the
	5	* information about the used sections of memory to
	6	* /sys/kernel/debug/kernel_pagetables.
	7	*
	8	* Derived from the arm64 implementation:
	9	* Copyright (c) 2014, The Linux Foundation, Laura Abbott.
	10	* (C) Copyright 2008 Intel Corporation, Arjan van de Ven.
	11	*
	12	* This program is free software; you can redistribute it and/or
	13	* modify it under the terms of the GNU General Public License
	14	* as published by the Free Software Foundation; version 2
	15	* of the License.
	16	*/
	17	#include <linux/debugfs.h>
	18	#include <linux/fs.h>
	19	#include <linux/io.h>
	20	#include <linux/mm.h>
	21	#include <linux/sched.h>
	22	#include <linux/seq_file.h>
	23	#include <asm/fixmap.h>
	24	#include <asm/pgtable.h>
	25	#include <linux/const.h>
	26	#include <asm/page.h>
	27	#include <asm/pgalloc.h>
	28
	29	/*
	30	* To visualise what is happening,
	31	*
	32	* - PTRS_PER_P = how many entries there are in the corresponding P
	33	* - P**_SHIFT = how many bits of the address we use to index into the
	34	* corresponding P**
	35	* - P**_SIZE is how much memory we can access through the table - not the
	36	* size of the table itself.
	37	* P**={PGD, PUD, PMD, PTE}
	38	*
	39	*
	40	* Each entry of the PGD points to a PUD. Each entry of a PUD points to a
	41	* PMD. Each entry of a PMD points to a PTE. And every PTE entry points to
	42	* a page.
	43	*
	44	* In the case where there are only 3 levels, the PUD is folded into the
	45	* PGD: every PUD has only one entry which points to the PMD.
	46	*
	47	* The page dumper groups page table entries of the same type into a single
	48	* description. It uses pg_state to track the range information while
	49	* iterating over the PTE entries. When the continuity is broken it then
	50	* dumps out a description of the range - ie PTEs that are virtually contiguous
	51	* with the same PTE flags are chunked together. This is to make it clear how
	52	* different areas of the kernel virtual memory are used.
	53	*
	54	*/
	55	struct pg_state {
	56	struct seq_file *seq;
	57	const struct addr_marker *marker;
	58	unsigned long start_address;
	59	unsigned int level;
	60	u64 current_flags;
	61	};
	62
	63	struct addr_marker {
	64	unsigned long start_address;
65	const char *name;
66	};
67
68	static struct addr_marker address_markers[] = {
69	{ 0, "Start of kernel VM" },
70	{ 0, "vmalloc() Area" },
71	{ 0, "vmalloc() End" },
72	{ 0, "isa I/O start" },
73	{ 0, "isa I/O end" },
74	{ 0, "phb I/O start" },
75	{ 0, "phb I/O end" },
76	{ 0, "I/O remap start" },
77	{ 0, "I/O remap end" },
78	{ 0, "vmemmap start" },
79	{ -1, NULL },
80	};
81
82	struct flag_info {
83	u64 mask;
84	u64 val;
85	const char *set;
86	const char *clear;
87	bool is_val;
88	int shift;
89	};
90
91	static const struct flag_info flag_array[] = {
92	{
93	#ifdef CONFIG_PPC_STD_MMU_64
94	.mask = _PAGE_PRIVILEGED,
95	.val = 0,
96	#else
97	.mask = _PAGE_USER,
98	.val = _PAGE_USER,
99	#endif
100	.set = "user",
101	.clear = " ",
102	}, {
103	.mask = _PAGE_RW,
104	.val = _PAGE_RW,
105	.set = "rw",
106	.clear = "ro",
107	}, {
108	.mask = _PAGE_EXEC,
109	.val = _PAGE_EXEC,
110	.set = " X ",
111	.clear = " ",
112	}, {
113	.mask = _PAGE_PTE,
114	.val = _PAGE_PTE,
115	.set = "pte",
116	.clear = " ",
117	}, {
118	.mask = _PAGE_PRESENT,
119	.val = _PAGE_PRESENT,
120	.set = "present",
121	.clear = " ",
122	}, {
123	#ifdef CONFIG_PPC_STD_MMU_64
124	.mask = H_PAGE_HASHPTE,
125	.val = H_PAGE_HASHPTE,
126	#else
127	.mask = _PAGE_HASHPTE,
128	.val = _PAGE_HASHPTE,
129	#endif
130	.set = "hpte",
131	.clear = " ",
132	}, {
133	#ifndef CONFIG_PPC_STD_MMU_64
134	.mask = _PAGE_GUARDED,
135	.val = _PAGE_GUARDED,
136	.set = "guarded",
137	.clear = " ",
138	}, {
139	#endif
140	.mask = _PAGE_DIRTY,
141	.val = _PAGE_DIRTY,
142	.set = "dirty",
143	.clear = " ",
144	}, {
145	.mask = _PAGE_ACCESSED,
146	.val = _PAGE_ACCESSED,
147	.set = "accessed",
148	.clear = " ",
149	}, {
150	#ifndef CONFIG_PPC_STD_MMU_64
151	.mask = _PAGE_WRITETHRU,
152	.val = _PAGE_WRITETHRU,
153	.set = "write through",
154	.clear = " ",
155	}, {
156	#endif
157	.mask = _PAGE_NO_CACHE,
158	.val = _PAGE_NO_CACHE,
159	.set = "no cache",
160	.clear = " ",
161	}, {
dd5ac03e	162	#ifdef CONFIG_PPC_BOOK3S_64
8eb07b18 RG	163	.mask = H_PAGE_BUSY,
	164	.val = H_PAGE_BUSY,
	165	.set = "busy",
	166	}, {
	167	#ifdef CONFIG_PPC_64K_PAGES
	168	.mask = H_PAGE_COMBO,
	169	.val = H_PAGE_COMBO,
	170	.set = "combo",
	171	}, {
	172	.mask = H_PAGE_4K_PFN,
	173	.val = H_PAGE_4K_PFN,
	174	.set = "4K_pfn",
	175	}, {
	176	#endif
	177	.mask = H_PAGE_F_GIX,
	178	.val = H_PAGE_F_GIX,
	179	.set = "f_gix",
	180	.is_val = true,
	181	.shift = H_PAGE_F_GIX_SHIFT,
	182	}, {
	183	.mask = H_PAGE_F_SECOND,
	184	.val = H_PAGE_F_SECOND,
	185	.set = "f_second",
	186	}, {
dd5ac03e	187	#endif
8eb07b18 RG	188	.mask = _PAGE_SPECIAL,
	189	.val = _PAGE_SPECIAL,
	190	.set = "special",
	191	}
	192	};
	193
	194	struct pgtable_level {
	195	const struct flag_info *flag;
	196	size_t num;
	197	u64 mask;
	198	};
	199
	200	static struct pgtable_level pg_level[] = {
	201	{
	202	}, { /* pgd */
	203	.flag = flag_array,
	204	.num = ARRAY_SIZE(flag_array),
	205	}, { /* pud */
	206	.flag = flag_array,
	207	.num = ARRAY_SIZE(flag_array),
	208	}, { /* pmd */
	209	.flag = flag_array,
	210	.num = ARRAY_SIZE(flag_array),
	211	}, { /* pte */
	212	.flag = flag_array,
	213	.num = ARRAY_SIZE(flag_array),
	214	},
	215	};
	216
	217	static void dump_flag_info(struct pg_state *st, const struct flag_info
	218	*flag, u64 pte, int num)
	219	{
	220	unsigned int i;
	221
	222	for (i = 0; i < num; i++, flag++) {
	223	const char *s = NULL;
	224	u64 val;
	225
	226	/* flag not defined so don't check it */
	227	if (flag->mask == 0)
	228	continue;
	229	/* Some 'flags' are actually values */
	230	if (flag->is_val) {
	231	val = pte & flag->val;
	232	if (flag->shift)
	233	val = val >> flag->shift;
	234	seq_printf(st->seq, " %s:%llx", flag->set, val);
	235	} else {
	236	if ((pte & flag->mask) == flag->val)
	237	s = flag->set;
	238	else
	239	s = flag->clear;
	240	if (s)
	241	seq_printf(st->seq, " %s", s);
	242	}
	243	st->current_flags &= ~flag->mask;
	244	}
	245	if (st->current_flags != 0)
	246	seq_printf(st->seq, " unknown flags:%llx", st->current_flags);
	247	}
	248
	249	static void dump_addr(struct pg_state *st, unsigned long addr)
	250	{
	251	static const char units[] = "KMGTPE";
252	const char *unit = units;
253	unsigned long delta;
254
255	seq_printf(st->seq, "0x%016lx-0x%016lx ", st->start_address, addr-1);
256	delta = (addr - st->start_address) >> 10;
257	/* Work out what appropriate unit to use */
258	while (!(delta & 1023) && unit[1]) {
259	delta >>= 10;
260	unit++;
261	}
262	seq_printf(st->seq, "%9lu%c", delta, *unit);
263
264	}
265
266	static void note_page(struct pg_state *st, unsigned long addr,
267	unsigned int level, u64 val)
268	{
269	u64 flag = val & pg_level[level].mask;
270	/* At first no level is set */
271	if (!st->level) {
272	st->level = level;
273	st->current_flags = flag;
274	st->start_address = addr;
275	seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
276	/*
277	* Dump the section of virtual memory when:
278	* - the PTE flags from one entry to the next differs.
279	* - we change levels in the tree.
280	* - the address is in a different section of memory and is thus
281	* used for a different purpose, regardless of the flags.
282	*/
283	} else if (flag != st->current_flags \|\| level != st->level \|\|
284	addr >= st->marker[1].start_address) {
285
286	/* Check the PTE flags */
287	if (st->current_flags) {
288	dump_addr(st, addr);
289
290	/* Dump all the flags */
291	if (pg_level[st->level].flag)
292	dump_flag_info(st, pg_level[st->level].flag,
293	st->current_flags,
294	pg_level[st->level].num);
295
296	seq_puts(st->seq, "\n");
297	}
298
299	/*
300	* Address indicates we have passed the end of the
301	* current section of virtual memory
302	*/
303	while (addr >= st->marker[1].start_address) {
304	st->marker++;
305	seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
306	}
307	st->start_address = addr;
308	st->current_flags = flag;
309	st->level = level;
310	}
311	}
312
313	static void walk_pte(struct pg_state st, pmd_t pmd, unsigned long start)
314	{
315	pte_t *pte = pte_offset_kernel(pmd, 0);
316	unsigned long addr;
317	unsigned int i;
318
319	for (i = 0; i < PTRS_PER_PTE; i++, pte++) {
320	addr = start + i * PAGE_SIZE;
321	note_page(st, addr, 4, pte_val(*pte));
322
323	}
324	}
325
326	static void walk_pmd(struct pg_state st, pud_t pud, unsigned long start)
327	{
328	pmd_t *pmd = pmd_offset(pud, 0);
329	unsigned long addr;
330	unsigned int i;
331
332	for (i = 0; i < PTRS_PER_PMD; i++, pmd++) {
333	addr = start + i * PMD_SIZE;
334	if (!pmd_none(*pmd))
335	/* pmd exists */
336	walk_pte(st, pmd, addr);
337	else
338	note_page(st, addr, 3, pmd_val(*pmd));
339	}
340	}
341
342	static void walk_pud(struct pg_state st, pgd_t pgd, unsigned long start)
343	{
344	pud_t *pud = pud_offset(pgd, 0);
345	unsigned long addr;
346	unsigned int i;
347
348	for (i = 0; i < PTRS_PER_PUD; i++, pud++) {
349	addr = start + i * PUD_SIZE;
350	if (!pud_none(*pud))
351	/* pud exists */
352	walk_pmd(st, pud, addr);
353	else
354	note_page(st, addr, 2, pud_val(*pud));
355	}
356	}
357
358	static void walk_pagetables(struct pg_state *st)
359	{
360	pgd_t *pgd = pgd_offset_k(0UL);
361	unsigned int i;
362	unsigned long addr;
363
364	/*
365	* Traverse the linux pagetable structure and dump pages that are in
366	* the hash pagetable.
367	*/
368	for (i = 0; i < PTRS_PER_PGD; i++, pgd++) {
369	addr = KERN_VIRT_START + i * PGDIR_SIZE;
370	if (!pgd_none(*pgd))
371	/* pgd exists */
372	walk_pud(st, pgd, addr);
373	else
374	note_page(st, addr, 1, pgd_val(*pgd));
375	}
376	}
377
378	static void populate_markers(void)
379	{
380	address_markers[0].start_address = PAGE_OFFSET;
381	address_markers[1].start_address = VMALLOC_START;
382	address_markers[2].start_address = VMALLOC_END;
383	address_markers[3].start_address = ISA_IO_BASE;
384	address_markers[4].start_address = ISA_IO_END;
385	address_markers[5].start_address = PHB_IO_BASE;
386	address_markers[6].start_address = PHB_IO_END;
387	address_markers[7].start_address = IOREMAP_BASE;
388	address_markers[8].start_address = IOREMAP_END;
389	#ifdef CONFIG_PPC_STD_MMU_64
390	address_markers[9].start_address = H_VMEMMAP_BASE;
391	#else
392	address_markers[9].start_address = VMEMMAP_BASE;
393	#endif
394	}
395
396	static int ptdump_show(struct seq_file m, void v)
397	{
398	struct pg_state st = {
399	.seq = m,
400	.start_address = KERN_VIRT_START,
401	.marker = address_markers,
402	};
403	/* Traverse kernel page tables */
404	walk_pagetables(&st);
405	note_page(&st, 0, 0, 0);
406	return 0;
407	}
408
409
410	static int ptdump_open(struct inode inode, struct file file)
411	{
412	return single_open(file, ptdump_show, NULL);
413	}
414
415	static const struct file_operations ptdump_fops = {
416	.open = ptdump_open,
417	.read = seq_read,
418	.llseek = seq_lseek,
419	.release = single_release,
420	};
421
422	static void build_pgtable_complete_mask(void)
423	{
424	unsigned int i, j;
425
426	for (i = 0; i < ARRAY_SIZE(pg_level); i++)
427	if (pg_level[i].flag)
428	for (j = 0; j < pg_level[i].num; j++)
429	pg_level[i].mask \|= pg_level[i].flag[j].mask;
430	}
431
432	static int ptdump_init(void)
433	{
434	struct dentry *debugfs_file;
435
436	populate_markers();
437	build_pgtable_complete_mask();
438	debugfs_file = debugfs_create_file("kernel_pagetables", 0400, NULL,
439	NULL, &ptdump_fops);
440	return debugfs_file ? 0 : -ENOMEM;
441	}
442	device_initcall(ptdump_init);