[linux-2.6-block.git] / arch / arm64 / mm / dump.c

/*
 * Copyright (c) 2014, The Linux Foundation. All rights reserved.
 * Debug helper to dump the current kernel pagetables of the system
 * so that we can see what the various memory ranges are set to.
 *
 * Derived from x86 and arm implementation:
 * (C) Copyright 2008 Intel Corporation
 *
 * Author: Arjan van de Ven <arjan@linux.intel.com>
 *
 * 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/errno.h>
#include <linux/fs.h>
#include <linux/io.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/sched.h>
#include <linux/seq_file.h>

#include <asm/fixmap.h>
#include <asm/kasan.h>
#include <asm/memory.h>
#include <asm/pgtable.h>
#include <asm/pgtable-hwdef.h>

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

static const struct addr_marker address_markers[] = {
#ifdef CONFIG_KASAN
	{ KASAN_SHADOW_START,		"Kasan shadow start" },
	{ KASAN_SHADOW_END,		"Kasan shadow end" },
#endif
	{ MODULES_VADDR,		"Modules start" },
	{ MODULES_END,			"Modules end" },
	{ VMALLOC_START,		"vmalloc() Area" },
	{ VMALLOC_END,			"vmalloc() End" },
	{ FIXADDR_START,		"Fixmap start" },
	{ FIXADDR_TOP,			"Fixmap end" },
	{ PCI_IO_START,			"PCI I/O start" },
	{ PCI_IO_END,			"PCI I/O end" },
#ifdef CONFIG_SPARSEMEM_VMEMMAP
	{ VMEMMAP_START,		"vmemmap start" },
	{ VMEMMAP_START + VMEMMAP_SIZE,	"vmemmap end" },
#endif
	{ PAGE_OFFSET,			"Linear Mapping" },
	{ -1,				NULL },
};

/*
 * 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.
 */
struct pg_state {
	struct seq_file *seq;
	const struct addr_marker *marker;
	unsigned long start_address;
	unsigned level;
	u64 current_prot;
};

struct prot_bits {
	u64		mask;
	u64		val;
	const char	*set;
	const char	*clear;
};

static const struct prot_bits pte_bits[] = {
	{
		.mask	= PTE_VALID,
		.val	= PTE_VALID,
		.set	= " ",
		.clear	= "F",
	}, {
		.mask	= PTE_USER,
		.val	= PTE_USER,
		.set	= "USR",
		.clear	= "   ",
	}, {
		.mask	= PTE_RDONLY,
		.val	= PTE_RDONLY,
		.set	= "ro",
		.clear	= "RW",
	}, {
		.mask	= PTE_PXN,
		.val	= PTE_PXN,
		.set	= "NX",
		.clear	= "x ",
	}, {
		.mask	= PTE_SHARED,
		.val	= PTE_SHARED,
		.set	= "SHD",
		.clear	= "   ",
	}, {
		.mask	= PTE_AF,
		.val	= PTE_AF,
		.set	= "AF",
		.clear	= "  ",
	}, {
		.mask	= PTE_NG,
		.val	= PTE_NG,
		.set	= "NG",
		.clear	= "  ",
	}, {
		.mask	= PTE_CONT,
		.val	= PTE_CONT,
		.set	= "CON",
		.clear	= "   ",
	}, {
		.mask	= PTE_TABLE_BIT,
		.val	= PTE_TABLE_BIT,
		.set	= "   ",
		.clear	= "BLK",
	}, {
		.mask	= PTE_UXN,
		.val	= PTE_UXN,
		.set	= "UXN",
	}, {
		.mask	= PTE_ATTRINDX_MASK,
		.val	= PTE_ATTRINDX(MT_DEVICE_nGnRnE),
		.set	= "DEVICE/nGnRnE",
	}, {
		.mask	= PTE_ATTRINDX_MASK,
		.val	= PTE_ATTRINDX(MT_DEVICE_nGnRE),
		.set	= "DEVICE/nGnRE",
	}, {
		.mask	= PTE_ATTRINDX_MASK,
		.val	= PTE_ATTRINDX(MT_DEVICE_GRE),
		.set	= "DEVICE/GRE",
	}, {
		.mask	= PTE_ATTRINDX_MASK,
		.val	= PTE_ATTRINDX(MT_NORMAL_NC),
		.set	= "MEM/NORMAL-NC",
	}, {
		.mask	= PTE_ATTRINDX_MASK,
		.val	= PTE_ATTRINDX(MT_NORMAL),
		.set	= "MEM/NORMAL",
	}
};

struct pg_level {
	const struct prot_bits *bits;
	const char *name;
	size_t num;
	u64 mask;
};

static struct pg_level pg_level[] = {
	{
	}, { /* pgd */
		.name	= "PGD",
		.bits	= pte_bits,
		.num	= ARRAY_SIZE(pte_bits),
	}, { /* pud */
		.name	= (CONFIG_PGTABLE_LEVELS > 3) ? "PUD" : "PGD",
		.bits	= pte_bits,
		.num	= ARRAY_SIZE(pte_bits),
	}, { /* pmd */
		.name	= (CONFIG_PGTABLE_LEVELS > 2) ? "PMD" : "PGD",
		.bits	= pte_bits,
		.num	= ARRAY_SIZE(pte_bits),
	}, { /* pte */
		.name	= "PTE",
		.bits	= pte_bits,
		.num	= ARRAY_SIZE(pte_bits),
	},
};

static void dump_prot(struct pg_state *st, const struct prot_bits *bits,
			size_t num)
{
	unsigned i;

	for (i = 0; i < num; i++, bits++) {
		const char *s;

		if ((st->current_prot & bits->mask) == bits->val)
			s = bits->set;
		else
			s = bits->clear;

		if (s)
			seq_printf(st->seq, " %s", s);
	}
}

static void note_page(struct pg_state *st, unsigned long addr, unsigned level,
				u64 val)
{
	static const char units[] = "KMGTPE";
	u64 prot = val & pg_level[level].mask;

	if (!st->level) {
		st->level = level;
		st->current_prot = prot;
		st->start_address = addr;
		seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
	} else if (prot != st->current_prot || level != st->level ||
		   addr >= st->marker[1].start_address) {
		const char *unit = units;
		unsigned long delta;

		if (st->current_prot) {
			seq_printf(st->seq, "0x%016lx-0x%016lx   ",
				   st->start_address, addr);

			delta = (addr - st->start_address) >> 10;
			while (!(delta & 1023) && unit[1]) {
				delta >>= 10;
				unit++;
			}
			seq_printf(st->seq, "%9lu%c %s", delta, *unit,
				   pg_level[st->level].name);
			if (pg_level[st->level].bits)
				dump_prot(st, pg_level[st->level].bits,
					  pg_level[st->level].num);
			seq_puts(st->seq, "\n");
		}

		if (addr >= st->marker[1].start_address) {
			st->marker++;
			seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
		}

		st->start_address = addr;
		st->current_prot = prot;
		st->level = level;
	}

	if (addr >= st->marker[1].start_address) {
		st->marker++;
		seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
	}

}

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 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 i;

	for (i = 0; i < PTRS_PER_PMD; i++, pmd++) {
		addr = start + i * PMD_SIZE;
		if (pmd_none(*pmd) || pmd_sect(*pmd)) {
			note_page(st, addr, 3, pmd_val(*pmd));
		} else {
			BUG_ON(pmd_bad(*pmd));
			walk_pte(st, pmd, addr);
		}
	}
}

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 i;

	for (i = 0; i < PTRS_PER_PUD; i++, pud++) {
		addr = start + i * PUD_SIZE;
		if (pud_none(*pud) || pud_sect(*pud)) {
			note_page(st, addr, 2, pud_val(*pud));
		} else {
			BUG_ON(pud_bad(*pud));
			walk_pmd(st, pud, addr);
		}
	}
}

static void walk_pgd(struct pg_state *st, struct mm_struct *mm, unsigned long start)
{
	pgd_t *pgd = pgd_offset(mm, 0UL);
	unsigned i;
	unsigned long addr;

	for (i = 0; i < PTRS_PER_PGD; i++, pgd++) {
		addr = start + i * PGDIR_SIZE;
		if (pgd_none(*pgd)) {
			note_page(st, addr, 1, pgd_val(*pgd));
		} else {
			BUG_ON(pgd_bad(*pgd));
			walk_pud(st, pgd, addr);
		}
	}
}

static int ptdump_show(struct seq_file *m, void *v)
{
	struct pg_state st = {
		.seq = m,
		.marker = address_markers,
	};

	walk_pgd(&st, &init_mm, VA_START);

	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 int ptdump_init(void)
{
	struct dentry *pe;
	unsigned i, j;

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

	pe = debugfs_create_file("kernel_page_tables", 0400, NULL, NULL,
				 &ptdump_fops);
	return pe ? 0 : -ENOMEM;
}
device_initcall(ptdump_init);
Commit	Line	Data
c9465b4e LA	1	/*
	2	* Copyright (c) 2014, The Linux Foundation. All rights reserved.
	3	* Debug helper to dump the current kernel pagetables of the system
	4	* so that we can see what the various memory ranges are set to.
	5	*
	6	* Derived from x86 and arm implementation:
	7	* (C) Copyright 2008 Intel Corporation
	8	*
	9	* Author: Arjan van de Ven <arjan@linux.intel.com>
	10	*
	11	* This program is free software; you can redistribute it and/or
	12	* modify it under the terms of the GNU General Public License
	13	* as published by the Free Software Foundation; version 2
	14	* of the License.
	15	*/
	16	#include <linux/debugfs.h>
764011ca	17	#include <linux/errno.h>
c9465b4e	18	#include <linux/fs.h>
284be285	19	#include <linux/io.h>
764011ca	20	#include <linux/init.h>
c9465b4e LA	21	#include <linux/mm.h>
	22	#include <linux/sched.h>
	23	#include <linux/seq_file.h>
	24
	25	#include <asm/fixmap.h>
d8fc68a0	26	#include <asm/kasan.h>
764011ca	27	#include <asm/memory.h>
c9465b4e	28	#include <asm/pgtable.h>
764011ca	29	#include <asm/pgtable-hwdef.h>
c9465b4e	30
c9465b4e LA	31	struct addr_marker {
	32	unsigned long start_address;
	33	const char *name;
	34	};
	35
c8f8cca4	36	static const struct addr_marker address_markers[] = {
d8fc68a0 AB	37	#ifdef CONFIG_KASAN
	38	{ KASAN_SHADOW_START, "Kasan shadow start" },
	39	{ KASAN_SHADOW_END, "Kasan shadow end" },
	40	#endif
c8f8cca4 AB	41	{ MODULES_VADDR, "Modules start" },
	42	{ MODULES_END, "Modules end" },
	43	{ VMALLOC_START, "vmalloc() Area" },
	44	{ VMALLOC_END, "vmalloc() End" },
	45	{ FIXADDR_START, "Fixmap start" },
	46	{ FIXADDR_TOP, "Fixmap end" },
	47	{ PCI_IO_START, "PCI I/O start" },
	48	{ PCI_IO_END, "PCI I/O end" },
3e1907d5	49	#ifdef CONFIG_SPARSEMEM_VMEMMAP
c8f8cca4 AB	50	{ VMEMMAP_START, "vmemmap start" },
c8f8cca4 AB	51	{ VMEMMAP_START + VMEMMAP_SIZE, "vmemmap end" },
3e1907d5	52	#endif
c8f8cca4 AB	53	{ PAGE_OFFSET, "Linear Mapping" },
c8f8cca4 AB	54	{ -1, NULL },
c9465b4e LA	55	};
c9465b4e LA	56
202e41a1 JL	57	/*
	58	* The page dumper groups page table entries of the same type into a single
	59	* description. It uses pg_state to track the range information while
	60	* iterating over the pte entries. When the continuity is broken it then
	61	* dumps out a description of the range.
	62	*/
c9465b4e LA	63	struct pg_state {
	64	struct seq_file *seq;
	65	const struct addr_marker *marker;
	66	unsigned long start_address;
	67	unsigned level;
	68	u64 current_prot;
	69	};
	70
	71	struct prot_bits {
	72	u64 mask;
	73	u64 val;
	74	const char *set;
	75	const char *clear;
	76	};
	77
	78	static const struct prot_bits pte_bits[] = {
	79	{
d7e9d594 LA	80	.mask = PTE_VALID,
	81	.val = PTE_VALID,
	82	.set = " ",
	83	.clear = "F",
	84	}, {
c9465b4e LA	85	.mask = PTE_USER,
	86	.val = PTE_USER,
	87	.set = "USR",
	88	.clear = " ",
	89	}, {
	90	.mask = PTE_RDONLY,
	91	.val = PTE_RDONLY,
	92	.set = "ro",
	93	.clear = "RW",
	94	}, {
	95	.mask = PTE_PXN,
	96	.val = PTE_PXN,
	97	.set = "NX",
	98	.clear = "x ",
	99	}, {
	100	.mask = PTE_SHARED,
	101	.val = PTE_SHARED,
	102	.set = "SHD",
	103	.clear = " ",
	104	}, {
	105	.mask = PTE_AF,
	106	.val = PTE_AF,
	107	.set = "AF",
	108	.clear = " ",
	109	}, {
	110	.mask = PTE_NG,
	111	.val = PTE_NG,
	112	.set = "NG",
	113	.clear = " ",
202e41a1 JL	114	}, {
	115	.mask = PTE_CONT,
	116	.val = PTE_CONT,
	117	.set = "CON",
	118	.clear = " ",
	119	}, {
	120	.mask = PTE_TABLE_BIT,
	121	.val = PTE_TABLE_BIT,
	122	.set = " ",
	123	.clear = "BLK",
c9465b4e LA	124	}, {
	125	.mask = PTE_UXN,
	126	.val = PTE_UXN,
	127	.set = "UXN",
	128	}, {
	129	.mask = PTE_ATTRINDX_MASK,
	130	.val = PTE_ATTRINDX(MT_DEVICE_nGnRnE),
	131	.set = "DEVICE/nGnRnE",
	132	}, {
	133	.mask = PTE_ATTRINDX_MASK,
	134	.val = PTE_ATTRINDX(MT_DEVICE_nGnRE),
	135	.set = "DEVICE/nGnRE",
	136	}, {
	137	.mask = PTE_ATTRINDX_MASK,
	138	.val = PTE_ATTRINDX(MT_DEVICE_GRE),
	139	.set = "DEVICE/GRE",
	140	}, {
	141	.mask = PTE_ATTRINDX_MASK,
	142	.val = PTE_ATTRINDX(MT_NORMAL_NC),
	143	.set = "MEM/NORMAL-NC",
	144	}, {
	145	.mask = PTE_ATTRINDX_MASK,
	146	.val = PTE_ATTRINDX(MT_NORMAL),
	147	.set = "MEM/NORMAL",
	148	}
	149	};
	150
	151	struct pg_level {
	152	const struct prot_bits *bits;
48dd73c5	153	const char *name;
c9465b4e LA	154	size_t num;
	155	u64 mask;
	156	};
	157
	158	static struct pg_level pg_level[] = {
	159	{
	160	}, { /* pgd */
48dd73c5	161	.name = "PGD",
c9465b4e LA	162	.bits = pte_bits,
	163	.num = ARRAY_SIZE(pte_bits),
	164	}, { /* pud */
48dd73c5	165	.name = (CONFIG_PGTABLE_LEVELS > 3) ? "PUD" : "PGD",
c9465b4e LA	166	.bits = pte_bits,
	167	.num = ARRAY_SIZE(pte_bits),
	168	}, { /* pmd */
48dd73c5	169	.name = (CONFIG_PGTABLE_LEVELS > 2) ? "PMD" : "PGD",
c9465b4e LA	170	.bits = pte_bits,
	171	.num = ARRAY_SIZE(pte_bits),
	172	}, { /* pte */
48dd73c5	173	.name = "PTE",
c9465b4e LA	174	.bits = pte_bits,
	175	.num = ARRAY_SIZE(pte_bits),
	176	},
	177	};
	178
	179	static void dump_prot(struct pg_state st, const struct prot_bits bits,
	180	size_t num)
	181	{
	182	unsigned i;
	183
	184	for (i = 0; i < num; i++, bits++) {
	185	const char *s;
	186
	187	if ((st->current_prot & bits->mask) == bits->val)
	188	s = bits->set;
	189	else
	190	s = bits->clear;
	191
	192	if (s)
	193	seq_printf(st->seq, " %s", s);
	194	}
	195	}
	196
	197	static void note_page(struct pg_state *st, unsigned long addr, unsigned level,
	198	u64 val)
	199	{
	200	static const char units[] = "KMGTPE";
	201	u64 prot = val & pg_level[level].mask;
	202
c9465b4e LA	203	if (!st->level) {
	204	st->level = level;
	205	st->current_prot = prot;
	206	st->start_address = addr;
	207	seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
	208	} else if (prot != st->current_prot \|\| level != st->level \|\|
	209	addr >= st->marker[1].start_address) {
	210	const char *unit = units;
	211	unsigned long delta;
	212
	213	if (st->current_prot) {
202e41a1	214	seq_printf(st->seq, "0x%016lx-0x%016lx ",
c9465b4e LA	215	st->start_address, addr);
	216
	217	delta = (addr - st->start_address) >> 10;
	218	while (!(delta & 1023) && unit[1]) {
	219	delta >>= 10;
	220	unit++;
	221	}
48dd73c5 MR	222	seq_printf(st->seq, "%9lu%c %s", delta, *unit,
48dd73c5 MR	223	pg_level[st->level].name);
c9465b4e LA	224	if (pg_level[st->level].bits)
	225	dump_prot(st, pg_level[st->level].bits,
	226	pg_level[st->level].num);
	227	seq_puts(st->seq, "\n");
	228	}
	229
	230	if (addr >= st->marker[1].start_address) {
	231	st->marker++;
	232	seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
	233	}
	234
	235	st->start_address = addr;
	236	st->current_prot = prot;
	237	st->level = level;
	238	}
	239
	240	if (addr >= st->marker[1].start_address) {
	241	st->marker++;
	242	seq_printf(st->seq, "---[ %s ]---\n", st->marker->name);
	243	}
	244
	245	}
	246
	247	static void walk_pte(struct pg_state st, pmd_t pmd, unsigned long start)
	248	{
	249	pte_t *pte = pte_offset_kernel(pmd, 0);
	250	unsigned long addr;
	251	unsigned i;
	252
	253	for (i = 0; i < PTRS_PER_PTE; i++, pte++) {
	254	addr = start + i * PAGE_SIZE;
	255	note_page(st, addr, 4, pte_val(*pte));
	256	}
	257	}
	258
	259	static void walk_pmd(struct pg_state st, pud_t pud, unsigned long start)
	260	{
	261	pmd_t *pmd = pmd_offset(pud, 0);
	262	unsigned long addr;
	263	unsigned i;
	264
	265	for (i = 0; i < PTRS_PER_PMD; i++, pmd++) {
	266	addr = start + i * PMD_SIZE;
a1c76574	267	if (pmd_none(pmd) \|\| pmd_sect(pmd)) {
c9465b4e	268	note_page(st, addr, 3, pmd_val(*pmd));
a1c76574 MR	269	} else {
a1c76574 MR	270	BUG_ON(pmd_bad(*pmd));
c9465b4e	271	walk_pte(st, pmd, addr);
a1c76574	272	}
c9465b4e LA	273	}
	274	}
	275
	276	static void walk_pud(struct pg_state st, pgd_t pgd, unsigned long start)
	277	{
	278	pud_t *pud = pud_offset(pgd, 0);
	279	unsigned long addr;
	280	unsigned i;
	281
	282	for (i = 0; i < PTRS_PER_PUD; i++, pud++) {
	283	addr = start + i * PUD_SIZE;
a1c76574	284	if (pud_none(pud) \|\| pud_sect(pud)) {
c9465b4e	285	note_page(st, addr, 2, pud_val(*pud));
a1c76574 MR	286	} else {
a1c76574 MR	287	BUG_ON(pud_bad(*pud));
c9465b4e	288	walk_pmd(st, pud, addr);
a1c76574	289	}
c9465b4e LA	290	}
	291	}
	292
	293	static void walk_pgd(struct pg_state st, struct mm_struct mm, unsigned long start)
	294	{
35545f0c	295	pgd_t *pgd = pgd_offset(mm, 0UL);
c9465b4e LA	296	unsigned i;
	297	unsigned long addr;
	298
	299	for (i = 0; i < PTRS_PER_PGD; i++, pgd++) {
	300	addr = start + i * PGDIR_SIZE;
a1c76574	301	if (pgd_none(*pgd)) {
c9465b4e	302	note_page(st, addr, 1, pgd_val(*pgd));
a1c76574 MR	303	} else {
a1c76574 MR	304	BUG_ON(pgd_bad(*pgd));
c9465b4e	305	walk_pud(st, pgd, addr);
a1c76574	306	}
c9465b4e LA	307	}
	308	}
	309
	310	static int ptdump_show(struct seq_file m, void v)
	311	{
	312	struct pg_state st = {
	313	.seq = m,
	314	.marker = address_markers,
	315	};
	316
cc30e6b9	317	walk_pgd(&st, &init_mm, VA_START);
c9465b4e LA	318
	319	note_page(&st, 0, 0, 0);
	320	return 0;
	321	}
	322
	323	static int ptdump_open(struct inode inode, struct file file)
	324	{
	325	return single_open(file, ptdump_show, NULL);
	326	}
	327
	328	static const struct file_operations ptdump_fops = {
	329	.open = ptdump_open,
	330	.read = seq_read,
	331	.llseek = seq_lseek,
	332	.release = single_release,
	333	};
	334
	335	static int ptdump_init(void)
	336	{
	337	struct dentry *pe;
	338	unsigned i, j;
	339
	340	for (i = 0; i < ARRAY_SIZE(pg_level); i++)
	341	if (pg_level[i].bits)
	342	for (j = 0; j < pg_level[i].num; j++)
	343	pg_level[i].mask \|= pg_level[i].bits[j].mask;
	344
c9465b4e LA	345	pe = debugfs_create_file("kernel_page_tables", 0400, NULL, NULL,
	346	&ptdump_fops);
	347	return pe ? 0 : -ENOMEM;
	348	}
	349	device_initcall(ptdump_init);