page_owner.c 11.9 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
#include <linux/stackdepot.h>

12 13
#include "internal.h"

14 15 16 17 18 19
/*
 * TODO: teach PAGE_OWNER_STACK_DEPTH (__dump_page_owner and save_stack)
 * to use off stack temporal storage
 */
#define PAGE_OWNER_STACK_DEPTH (16)

20
static bool page_owner_disabled = true;
21
DEFINE_STATIC_KEY_FALSE(page_owner_inited);
22

23 24 25
static depot_stack_handle_t dummy_handle;
static depot_stack_handle_t failure_handle;

26 27
static void init_early_allocated_pages(void);

28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
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;
}

48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
static noinline void register_dummy_stack(void)
{
	unsigned long entries[4];
	struct stack_trace dummy;

	dummy.nr_entries = 0;
	dummy.max_entries = ARRAY_SIZE(entries);
	dummy.entries = &entries[0];
	dummy.skip = 0;

	save_stack_trace(&dummy);
	dummy_handle = depot_save_stack(&dummy, GFP_KERNEL);
}

static noinline void register_failure_stack(void)
{
	unsigned long entries[4];
	struct stack_trace failure;

	failure.nr_entries = 0;
	failure.max_entries = ARRAY_SIZE(entries);
	failure.entries = &entries[0];
	failure.skip = 0;

	save_stack_trace(&failure);
	failure_handle = depot_save_stack(&failure, GFP_KERNEL);
}

76 77 78 79 80
static void init_page_owner(void)
{
	if (page_owner_disabled)
		return;

81 82
	register_dummy_stack();
	register_failure_stack();
83
	static_branch_enable(&page_owner_inited);
84
	init_early_allocated_pages();
85 86 87 88 89 90 91 92 93 94 95 96 97 98
}

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);
99 100
		if (unlikely(!page_ext))
			continue;
101 102 103 104
		__clear_bit(PAGE_EXT_OWNER, &page_ext->flags);
	}
}

105 106
static inline bool check_recursive_alloc(struct stack_trace *trace,
					unsigned long ip)
107
{
108 109 110 111 112 113 114 115 116
	int i, count;

	if (!trace->nr_entries)
		return false;

	for (i = 0, count = 0; i < trace->nr_entries; i++) {
		if (trace->entries[i] == ip && ++count == 2)
			return true;
	}
117

118 119 120 121 122 123
	return false;
}

static noinline depot_stack_handle_t save_stack(gfp_t flags)
{
	unsigned long entries[PAGE_OWNER_STACK_DEPTH];
124 125
	struct stack_trace trace = {
		.nr_entries = 0,
126 127 128
		.entries = entries,
		.max_entries = PAGE_OWNER_STACK_DEPTH,
		.skip = 0
129
	};
130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
	depot_stack_handle_t handle;

	save_stack_trace(&trace);
	if (trace.nr_entries != 0 &&
	    trace.entries[trace.nr_entries-1] == ULONG_MAX)
		trace.nr_entries--;

	/*
	 * We need to check recursion here because our request to stackdepot
	 * could trigger memory allocation to save new entry. New memory
	 * allocation would reach here and call depot_save_stack() again
	 * if we don't catch it. There is still not enough memory in stackdepot
	 * so it would try to allocate memory again and loop forever.
	 */
	if (check_recursive_alloc(&trace, _RET_IP_))
		return dummy_handle;

	handle = depot_save_stack(&trace, flags);
	if (!handle)
		handle = failure_handle;

	return handle;
}

noinline void __set_page_owner(struct page *page, unsigned int order,
					gfp_t gfp_mask)
{
	struct page_ext *page_ext = lookup_page_ext(page);
158

159 160 161
	if (unlikely(!page_ext))
		return;

162
	page_ext->handle = save_stack(gfp_mask);
163 164
	page_ext->order = order;
	page_ext->gfp_mask = gfp_mask;
165
	page_ext->last_migrate_reason = -1;
166 167 168 169

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

170 171 172
void __set_page_owner_migrate_reason(struct page *page, int reason)
{
	struct page_ext *page_ext = lookup_page_ext(page);
173 174
	if (unlikely(!page_ext))
		return;
175 176 177 178

	page_ext->last_migrate_reason = reason;
}

179
void __split_page_owner(struct page *page, unsigned int order)
180
{
181
	int i;
182
	struct page_ext *page_ext = lookup_page_ext(page);
183

184
	if (unlikely(!page_ext))
185
		return;
186

187 188 189
	page_ext->order = 0;
	for (i = 1; i < (1 << order); i++)
		__copy_page_owner(page, page + i);
190 191
}

192 193 194 195 196
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);

197 198 199
	if (unlikely(!old_ext || !new_ext))
		return;

200 201
	new_ext->order = old_ext->order;
	new_ext->gfp_mask = old_ext->gfp_mask;
202
	new_ext->last_migrate_reason = old_ext->last_migrate_reason;
203
	new_ext->handle = old_ext->handle;
204 205 206 207 208 209 210 211 212 213 214 215 216

	/*
	 * 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);
}

217 218
static ssize_t
print_page_owner(char __user *buf, size_t count, unsigned long pfn,
219 220
		struct page *page, struct page_ext *page_ext,
		depot_stack_handle_t handle)
221 222 223 224
{
	int ret;
	int pageblock_mt, page_mt;
	char *kbuf;
225
	unsigned long entries[PAGE_OWNER_STACK_DEPTH];
226
	struct stack_trace trace = {
227 228 229 230
		.nr_entries = 0,
		.entries = entries,
		.max_entries = PAGE_OWNER_STACK_DEPTH,
		.skip = 0
231
	};
232 233 234 235 236 237

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

	ret = snprintf(kbuf, count,
238 239 240
			"Page allocated via order %u, mask %#x(%pGg)\n",
			page_ext->order, page_ext->gfp_mask,
			&page_ext->gfp_mask);
241 242 243 244 245

	if (ret >= count)
		goto err;

	/* Print information relevant to grouping pages by mobility */
246
	pageblock_mt = get_pageblock_migratetype(page);
247 248
	page_mt  = gfpflags_to_migratetype(page_ext->gfp_mask);
	ret += snprintf(kbuf + ret, count - ret,
249
			"PFN %lu type %s Block %lu type %s Flags %#lx(%pGp)\n",
250
			pfn,
251
			migratetype_names[page_mt],
252
			pfn >> pageblock_order,
253 254
			migratetype_names[pageblock_mt],
			page->flags, &page->flags);
255 256 257 258

	if (ret >= count)
		goto err;

259
	depot_fetch_stack(handle, &trace);
260
	ret += snprint_stack_trace(kbuf + ret, count - ret, &trace, 0);
261 262 263
	if (ret >= count)
		goto err;

264 265 266 267 268 269 270 271
	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;
	}

272 273 274 275 276 277 278 279 280 281 282 283 284 285 286
	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;
}

287 288 289
void __dump_page_owner(struct page *page)
{
	struct page_ext *page_ext = lookup_page_ext(page);
290
	unsigned long entries[PAGE_OWNER_STACK_DEPTH];
291
	struct stack_trace trace = {
292 293 294 295
		.nr_entries = 0,
		.entries = entries,
		.max_entries = PAGE_OWNER_STACK_DEPTH,
		.skip = 0
296
	};
297
	depot_stack_handle_t handle;
298 299
	gfp_t gfp_mask;
	int mt;
300

301 302 303 304
	if (unlikely(!page_ext)) {
		pr_alert("There is not page extension available.\n");
		return;
	}
305 306
	gfp_mask = page_ext->gfp_mask;
	mt = gfpflags_to_migratetype(gfp_mask);
307

308 309 310 311 312
	if (!test_bit(PAGE_EXT_OWNER, &page_ext->flags)) {
		pr_alert("page_owner info is not active (free page?)\n");
		return;
	}

313 314 315 316 317 318 319
	handle = READ_ONCE(page_ext->handle);
	if (!handle) {
		pr_alert("page_owner info is not active (free page?)\n");
		return;
	}

	depot_fetch_stack(handle, &trace);
Joe Perches's avatar
Joe Perches committed
320 321
	pr_alert("page allocated via order %u, migratetype %s, gfp_mask %#x(%pGg)\n",
		 page_ext->order, migratetype_names[mt], gfp_mask, &gfp_mask);
322 323 324 325 326 327 328
	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]);
}

329 330 331 332 333 334
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;
335
	depot_stack_handle_t handle;
336

337
	if (!static_branch_unlikely(&page_owner_inited))
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
		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);
374 375
		if (unlikely(!page_ext))
			continue;
376 377

		/*
378 379
		 * Some pages could be missed by concurrent allocation or free,
		 * because we don't hold the zone lock.
380 381 382 383
		 */
		if (!test_bit(PAGE_EXT_OWNER, &page_ext->flags))
			continue;

384 385 386 387 388 389 390 391
		/*
		 * Access to page_ext->handle isn't synchronous so we should
		 * be careful to access it.
		 */
		handle = READ_ONCE(page_ext->handle);
		if (!handle)
			continue;

392 393 394
		/* Record the next PFN to read in the file offset */
		*ppos = (pfn - min_low_pfn) + 1;

395 396
		return print_page_owner(buf, count, pfn, page,
				page_ext, handle);
397 398 399 400 401
	}

	return 0;
}

402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
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);

435 436 437
			if (page_zone(page) != zone)
				continue;

438 439 440 441 442 443 444 445 446 447 448 449 450
			/*
			 * 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);
451 452
			if (unlikely(!page_ext))
				continue;
453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492

			/* 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);
}

493 494 495 496 497 498 499 500
static const struct file_operations proc_page_owner_operations = {
	.read		= read_page_owner,
};

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

501
	if (!static_branch_unlikely(&page_owner_inited)) {
502 503 504 505 506 507 508 509 510 511 512
		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;
}
513
late_initcall(pageowner_init)