page_owner.c 9.42 KB
Newer Older
1 2 3 4 5 6 7
#include <linux/debugfs.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/uaccess.h>
#include <linux/bootmem.h>
#include <linux/stacktrace.h>
#include <linux/page_owner.h>
8
#include <linux/jump_label.h>
9
#include <linux/migrate.h>
10 11 12
#include "internal.h"

static bool page_owner_disabled = true;
13
DEFINE_STATIC_KEY_FALSE(page_owner_inited);
14

15 16
static void init_early_allocated_pages(void);

17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
static int early_page_owner_param(char *buf)
{
	if (!buf)
		return -EINVAL;

	if (strcmp(buf, "on") == 0)
		page_owner_disabled = false;

	return 0;
}
early_param("page_owner", early_page_owner_param);

static bool need_page_owner(void)
{
	if (page_owner_disabled)
		return false;

	return true;
}

static void init_page_owner(void)
{
	if (page_owner_disabled)
		return;

42
	static_branch_enable(&page_owner_inited);
43
	init_early_allocated_pages();
44 45 46 47 48 49 50 51 52 53 54 55 56 57
}

struct page_ext_operations page_owner_ops = {
	.need = need_page_owner,
	.init = init_page_owner,
};

void __reset_page_owner(struct page *page, unsigned int order)
{
	int i;
	struct page_ext *page_ext;

	for (i = 0; i < (1 << order); i++) {
		page_ext = lookup_page_ext(page + i);
58 59
		if (unlikely(!page_ext))
			continue;
60 61 62 63 64 65
		__clear_bit(PAGE_EXT_OWNER, &page_ext->flags);
	}
}

void __set_page_owner(struct page *page, unsigned int order, gfp_t gfp_mask)
{
66
	struct page_ext *page_ext = lookup_page_ext(page);
67

68 69 70 71 72 73
	struct stack_trace trace = {
		.nr_entries = 0,
		.max_entries = ARRAY_SIZE(page_ext->trace_entries),
		.entries = &page_ext->trace_entries[0],
		.skip = 3,
	};
74

75 76 77
	if (unlikely(!page_ext))
		return;

78
	save_stack_trace(&trace);
79 80 81

	page_ext->order = order;
	page_ext->gfp_mask = gfp_mask;
82
	page_ext->nr_entries = trace.nr_entries;
83
	page_ext->last_migrate_reason = -1;
84 85 86 87

	__set_bit(PAGE_EXT_OWNER, &page_ext->flags);
}

88 89 90
void __set_page_owner_migrate_reason(struct page *page, int reason)
{
	struct page_ext *page_ext = lookup_page_ext(page);
91 92
	if (unlikely(!page_ext))
		return;
93 94 95 96

	page_ext->last_migrate_reason = reason;
}

97 98 99
gfp_t __get_page_owner_gfp(struct page *page)
{
	struct page_ext *page_ext = lookup_page_ext(page);
100 101 102 103 104 105
	if (unlikely(!page_ext))
		/*
		 * The caller just returns 0 if no valid gfp
		 * So return 0 here too.
		 */
		return 0;
106 107 108 109

	return page_ext->gfp_mask;
}

110 111 112 113 114 115
void __copy_page_owner(struct page *oldpage, struct page *newpage)
{
	struct page_ext *old_ext = lookup_page_ext(oldpage);
	struct page_ext *new_ext = lookup_page_ext(newpage);
	int i;

116 117 118
	if (unlikely(!old_ext || !new_ext))
		return;

119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
	new_ext->order = old_ext->order;
	new_ext->gfp_mask = old_ext->gfp_mask;
	new_ext->nr_entries = old_ext->nr_entries;

	for (i = 0; i < ARRAY_SIZE(new_ext->trace_entries); i++)
		new_ext->trace_entries[i] = old_ext->trace_entries[i];

	/*
	 * We don't clear the bit on the oldpage as it's going to be freed
	 * after migration. Until then, the info can be useful in case of
	 * a bug, and the overal stats will be off a bit only temporarily.
	 * Also, migrate_misplaced_transhuge_page() can still fail the
	 * migration and then we want the oldpage to retain the info. But
	 * in that case we also don't need to explicitly clear the info from
	 * the new page, which will be freed.
	 */
	__set_bit(PAGE_EXT_OWNER, &new_ext->flags);
}

138 139 140 141 142 143 144
static ssize_t
print_page_owner(char __user *buf, size_t count, unsigned long pfn,
		struct page *page, struct page_ext *page_ext)
{
	int ret;
	int pageblock_mt, page_mt;
	char *kbuf;
145 146 147 148
	struct stack_trace trace = {
		.nr_entries = page_ext->nr_entries,
		.entries = &page_ext->trace_entries[0],
	};
149 150 151 152 153 154

	kbuf = kmalloc(count, GFP_KERNEL);
	if (!kbuf)
		return -ENOMEM;

	ret = snprintf(kbuf, count,
155 156 157
			"Page allocated via order %u, mask %#x(%pGg)\n",
			page_ext->order, page_ext->gfp_mask,
			&page_ext->gfp_mask);
158 159 160 161 162

	if (ret >= count)
		goto err;

	/* Print information relevant to grouping pages by mobility */
163
	pageblock_mt = get_pageblock_migratetype(page);
164 165
	page_mt  = gfpflags_to_migratetype(page_ext->gfp_mask);
	ret += snprintf(kbuf + ret, count - ret,
166
			"PFN %lu type %s Block %lu type %s Flags %#lx(%pGp)\n",
167
			pfn,
168
			migratetype_names[page_mt],
169
			pfn >> pageblock_order,
170 171
			migratetype_names[pageblock_mt],
			page->flags, &page->flags);
172 173 174 175

	if (ret >= count)
		goto err;

176
	ret += snprint_stack_trace(kbuf + ret, count - ret, &trace, 0);
177 178 179
	if (ret >= count)
		goto err;

180 181 182 183 184 185 186 187
	if (page_ext->last_migrate_reason != -1) {
		ret += snprintf(kbuf + ret, count - ret,
			"Page has been migrated, last migrate reason: %s\n",
			migrate_reason_names[page_ext->last_migrate_reason]);
		if (ret >= count)
			goto err;
	}

188 189 190 191 192 193 194 195 196 197 198 199 200 201 202
	ret += snprintf(kbuf + ret, count - ret, "\n");
	if (ret >= count)
		goto err;

	if (copy_to_user(buf, kbuf, ret))
		ret = -EFAULT;

	kfree(kbuf);
	return ret;

err:
	kfree(kbuf);
	return -ENOMEM;
}

203 204 205 206 207 208 209 210 211 212
void __dump_page_owner(struct page *page)
{
	struct page_ext *page_ext = lookup_page_ext(page);
	struct stack_trace trace = {
		.nr_entries = page_ext->nr_entries,
		.entries = &page_ext->trace_entries[0],
	};
	gfp_t gfp_mask = page_ext->gfp_mask;
	int mt = gfpflags_to_migratetype(gfp_mask);

213 214 215 216 217
	if (unlikely(!page_ext)) {
		pr_alert("There is not page extension available.\n");
		return;
	}

218 219 220 221 222
	if (!test_bit(PAGE_EXT_OWNER, &page_ext->flags)) {
		pr_alert("page_owner info is not active (free page?)\n");
		return;
	}

Joe Perches's avatar
Joe Perches committed
223 224
	pr_alert("page allocated via order %u, migratetype %s, gfp_mask %#x(%pGg)\n",
		 page_ext->order, migratetype_names[mt], gfp_mask, &gfp_mask);
225 226 227 228 229 230 231
	print_stack_trace(&trace, 0);

	if (page_ext->last_migrate_reason != -1)
		pr_alert("page has been migrated, last migrate reason: %s\n",
			migrate_reason_names[page_ext->last_migrate_reason]);
}

232 233 234 235 236 237 238
static ssize_t
read_page_owner(struct file *file, char __user *buf, size_t count, loff_t *ppos)
{
	unsigned long pfn;
	struct page *page;
	struct page_ext *page_ext;

239
	if (!static_branch_unlikely(&page_owner_inited))
240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
		return -EINVAL;

	page = NULL;
	pfn = min_low_pfn + *ppos;

	/* Find a valid PFN or the start of a MAX_ORDER_NR_PAGES area */
	while (!pfn_valid(pfn) && (pfn & (MAX_ORDER_NR_PAGES - 1)) != 0)
		pfn++;

	drain_all_pages(NULL);

	/* Find an allocated page */
	for (; pfn < max_pfn; pfn++) {
		/*
		 * If the new page is in a new MAX_ORDER_NR_PAGES area,
		 * validate the area as existing, skip it if not
		 */
		if ((pfn & (MAX_ORDER_NR_PAGES - 1)) == 0 && !pfn_valid(pfn)) {
			pfn += MAX_ORDER_NR_PAGES - 1;
			continue;
		}

		/* Check for holes within a MAX_ORDER area */
		if (!pfn_valid_within(pfn))
			continue;

		page = pfn_to_page(pfn);
		if (PageBuddy(page)) {
			unsigned long freepage_order = page_order_unsafe(page);

			if (freepage_order < MAX_ORDER)
				pfn += (1UL << freepage_order) - 1;
			continue;
		}

		page_ext = lookup_page_ext(page);
276 277
		if (unlikely(!page_ext))
			continue;
278 279

		/*
280 281
		 * Some pages could be missed by concurrent allocation or free,
		 * because we don't hold the zone lock.
282 283 284 285 286 287 288 289 290 291 292 293 294
		 */
		if (!test_bit(PAGE_EXT_OWNER, &page_ext->flags))
			continue;

		/* Record the next PFN to read in the file offset */
		*ppos = (pfn - min_low_pfn) + 1;

		return print_page_owner(buf, count, pfn, page, page_ext);
	}

	return 0;
}

295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
static void init_pages_in_zone(pg_data_t *pgdat, struct zone *zone)
{
	struct page *page;
	struct page_ext *page_ext;
	unsigned long pfn = zone->zone_start_pfn, block_end_pfn;
	unsigned long end_pfn = pfn + zone->spanned_pages;
	unsigned long count = 0;

	/* Scan block by block. First and last block may be incomplete */
	pfn = zone->zone_start_pfn;

	/*
	 * Walk the zone in pageblock_nr_pages steps. If a page block spans
	 * a zone boundary, it will be double counted between zones. This does
	 * not matter as the mixed block count will still be correct
	 */
	for (; pfn < end_pfn; ) {
		if (!pfn_valid(pfn)) {
			pfn = ALIGN(pfn + 1, MAX_ORDER_NR_PAGES);
			continue;
		}

		block_end_pfn = ALIGN(pfn + 1, pageblock_nr_pages);
		block_end_pfn = min(block_end_pfn, end_pfn);

		page = pfn_to_page(pfn);

		for (; pfn < block_end_pfn; pfn++) {
			if (!pfn_valid_within(pfn))
				continue;

			page = pfn_to_page(pfn);

328 329 330
			if (page_zone(page) != zone)
				continue;

331 332 333 334 335 336 337 338 339 340 341 342 343
			/*
			 * We are safe to check buddy flag and order, because
			 * this is init stage and only single thread runs.
			 */
			if (PageBuddy(page)) {
				pfn += (1UL << page_order(page)) - 1;
				continue;
			}

			if (PageReserved(page))
				continue;

			page_ext = lookup_page_ext(page);
344 345
			if (unlikely(!page_ext))
				continue;
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385

			/* Maybe overraping zone */
			if (test_bit(PAGE_EXT_OWNER, &page_ext->flags))
				continue;

			/* Found early allocated page */
			set_page_owner(page, 0, 0);
			count++;
		}
	}

	pr_info("Node %d, zone %8s: page owner found early allocated %lu pages\n",
		pgdat->node_id, zone->name, count);
}

static void init_zones_in_node(pg_data_t *pgdat)
{
	struct zone *zone;
	struct zone *node_zones = pgdat->node_zones;
	unsigned long flags;

	for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; ++zone) {
		if (!populated_zone(zone))
			continue;

		spin_lock_irqsave(&zone->lock, flags);
		init_pages_in_zone(pgdat, zone);
		spin_unlock_irqrestore(&zone->lock, flags);
	}
}

static void init_early_allocated_pages(void)
{
	pg_data_t *pgdat;

	drain_all_pages(NULL);
	for_each_online_pgdat(pgdat)
		init_zones_in_node(pgdat);
}

386 387 388 389 390 391 392 393
static const struct file_operations proc_page_owner_operations = {
	.read		= read_page_owner,
};

static int __init pageowner_init(void)
{
	struct dentry *dentry;

394
	if (!static_branch_unlikely(&page_owner_inited)) {
395 396 397 398 399 400 401 402 403 404 405
		pr_info("page_owner is disabled\n");
		return 0;
	}

	dentry = debugfs_create_file("page_owner", S_IRUSR, NULL,
			NULL, &proc_page_owner_operations);
	if (IS_ERR(dentry))
		return PTR_ERR(dentry);

	return 0;
}
406
late_initcall(pageowner_init)