mempolicy.c 50.4 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4
/*
 * Simple NUMA memory policy for the Linux kernel.
 *
 * Copyright 2003,2004 Andi Kleen, SuSE Labs.
5
 * (C) Copyright 2005 Christoph Lameter, Silicon Graphics, Inc.
L
Linus Torvalds 已提交
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 * Subject to the GNU Public License, version 2.
 *
 * NUMA policy allows the user to give hints in which node(s) memory should
 * be allocated.
 *
 * Support four policies per VMA and per process:
 *
 * The VMA policy has priority over the process policy for a page fault.
 *
 * interleave     Allocate memory interleaved over a set of nodes,
 *                with normal fallback if it fails.
 *                For VMA based allocations this interleaves based on the
 *                offset into the backing object or offset into the mapping
 *                for anonymous memory. For process policy an process counter
 *                is used.
21
 *
L
Linus Torvalds 已提交
22 23
 * bind           Only allocate memory on a specific set of nodes,
 *                no fallback.
24 25 26 27
 *                FIXME: memory is allocated starting with the first node
 *                to the last. It would be better if bind would truly restrict
 *                the allocation to memory nodes instead
 *
L
Linus Torvalds 已提交
28 29 30 31 32
 * preferred       Try a specific node first before normal fallback.
 *                As a special case node -1 here means do the allocation
 *                on the local CPU. This is normally identical to default,
 *                but useful to set in a VMA when you have a non default
 *                process policy.
33
 *
L
Linus Torvalds 已提交
34 35 36 37 38 39 40 41 42 43 44 45 46 47 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 76 77 78 79 80 81 82 83
 * default        Allocate on the local node first, or when on a VMA
 *                use the process policy. This is what Linux always did
 *		  in a NUMA aware kernel and still does by, ahem, default.
 *
 * The process policy is applied for most non interrupt memory allocations
 * in that process' context. Interrupts ignore the policies and always
 * try to allocate on the local CPU. The VMA policy is only applied for memory
 * allocations for a VMA in the VM.
 *
 * Currently there are a few corner cases in swapping where the policy
 * is not applied, but the majority should be handled. When process policy
 * is used it is not remembered over swap outs/swap ins.
 *
 * Only the highest zone in the zone hierarchy gets policied. Allocations
 * requesting a lower zone just use default policy. This implies that
 * on systems with highmem kernel lowmem allocation don't get policied.
 * Same with GFP_DMA allocations.
 *
 * For shmfs/tmpfs/hugetlbfs shared memory the policy is shared between
 * all users and remembered even when nobody has memory mapped.
 */

/* Notebook:
   fix mmap readahead to honour policy and enable policy for any page cache
   object
   statistics for bigpages
   global policy for page cache? currently it uses process policy. Requires
   first item above.
   handle mremap for shared memory (currently ignored for the policy)
   grows down?
   make bind policy root only? It can trigger oom much faster and the
   kernel is not always grateful with that.
   could replace all the switch()es with a mempolicy_ops structure.
*/

#include <linux/mempolicy.h>
#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/hugetlb.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/nodemask.h>
#include <linux/cpuset.h>
#include <linux/gfp.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/module.h>
#include <linux/interrupt.h>
#include <linux/init.h>
#include <linux/compat.h>
84
#include <linux/swap.h>
85 86
#include <linux/seq_file.h>
#include <linux/proc_fs.h>
C
Christoph Lameter 已提交
87
#include <linux/migrate.h>
88
#include <linux/rmap.h>
89
#include <linux/security.h>
90

L
Linus Torvalds 已提交
91 92 93
#include <asm/tlbflush.h>
#include <asm/uaccess.h>

94
/* Internal flags */
95
#define MPOL_MF_DISCONTIG_OK (MPOL_MF_INTERNAL << 0)	/* Skip checks for continuous vmas */
96
#define MPOL_MF_INVERT (MPOL_MF_INTERNAL << 1)		/* Invert check for nodemask */
97
#define MPOL_MF_STATS (MPOL_MF_INTERNAL << 2)		/* Gather statistics */
98

99 100
static struct kmem_cache *policy_cache;
static struct kmem_cache *sn_cache;
L
Linus Torvalds 已提交
101 102 103

/* Highest zone. An specific allocation for a zone below that is not
   policied. */
104
enum zone_type policy_zone = 0;
L
Linus Torvalds 已提交
105

106
struct mempolicy default_policy = {
L
Linus Torvalds 已提交
107 108 109 110 111
	.refcnt = ATOMIC_INIT(1), /* never free it */
	.policy = MPOL_DEFAULT,
};

/* Do sanity checking on a policy */
112
static int mpol_check_policy(int mode, nodemask_t *nodes)
L
Linus Torvalds 已提交
113
{
114
	int empty = nodes_empty(*nodes);
L
Linus Torvalds 已提交
115 116 117 118 119 120 121 122 123 124 125 126 127 128

	switch (mode) {
	case MPOL_DEFAULT:
		if (!empty)
			return -EINVAL;
		break;
	case MPOL_BIND:
	case MPOL_INTERLEAVE:
		/* Preferred will only use the first bit, but allow
		   more for now. */
		if (empty)
			return -EINVAL;
		break;
	}
129
 	return nodes_subset(*nodes, node_states[N_HIGH_MEMORY]) ? 0 : -EINVAL;
L
Linus Torvalds 已提交
130
}
131

L
Linus Torvalds 已提交
132
/* Generate a custom zonelist for the BIND policy. */
133
static struct zonelist *bind_zonelist(nodemask_t *nodes)
L
Linus Torvalds 已提交
134 135
{
	struct zonelist *zl;
136 137
	int num, max, nd;
	enum zone_type k;
L
Linus Torvalds 已提交
138

139
	max = 1 + MAX_NR_ZONES * nodes_weight(*nodes);
140
	max++;			/* space for zlcache_ptr (see mmzone.h) */
141
	zl = kmalloc(sizeof(struct zone *) * max, GFP_KERNEL);
L
Linus Torvalds 已提交
142
	if (!zl)
143
		return ERR_PTR(-ENOMEM);
144
	zl->zlcache_ptr = NULL;
L
Linus Torvalds 已提交
145
	num = 0;
146 147 148 149
	/* First put in the highest zones from all nodes, then all the next 
	   lower zones etc. Avoid empty zones because the memory allocator
	   doesn't like them. If you implement node hot removal you
	   have to fix that. */
150
	k = MAX_NR_ZONES - 1;
151
	while (1) {
152 153 154 155 156
		for_each_node_mask(nd, *nodes) { 
			struct zone *z = &NODE_DATA(nd)->node_zones[k];
			if (z->present_pages > 0) 
				zl->zones[num++] = z;
		}
157 158 159
		if (k == 0)
			break;
		k--;
160
	}
161 162 163 164
	if (num == 0) {
		kfree(zl);
		return ERR_PTR(-EINVAL);
	}
L
Linus Torvalds 已提交
165 166 167 168 169
	zl->zones[num] = NULL;
	return zl;
}

/* Create a new policy */
170
static struct mempolicy *mpol_new(int mode, nodemask_t *nodes)
L
Linus Torvalds 已提交
171 172 173
{
	struct mempolicy *policy;

174 175 176
	pr_debug("setting mode %d nodes[0] %lx\n",
		 mode, nodes ? nodes_addr(*nodes)[0] : -1);

L
Linus Torvalds 已提交
177 178 179 180 181 182 183 184
	if (mode == MPOL_DEFAULT)
		return NULL;
	policy = kmem_cache_alloc(policy_cache, GFP_KERNEL);
	if (!policy)
		return ERR_PTR(-ENOMEM);
	atomic_set(&policy->refcnt, 1);
	switch (mode) {
	case MPOL_INTERLEAVE:
185
		policy->v.nodes = *nodes;
186 187 188
		nodes_and(policy->v.nodes, policy->v.nodes,
					node_states[N_HIGH_MEMORY]);
		if (nodes_weight(policy->v.nodes) == 0) {
189 190 191
			kmem_cache_free(policy_cache, policy);
			return ERR_PTR(-EINVAL);
		}
L
Linus Torvalds 已提交
192 193
		break;
	case MPOL_PREFERRED:
194
		policy->v.preferred_node = first_node(*nodes);
L
Linus Torvalds 已提交
195 196 197 198 199
		if (policy->v.preferred_node >= MAX_NUMNODES)
			policy->v.preferred_node = -1;
		break;
	case MPOL_BIND:
		policy->v.zonelist = bind_zonelist(nodes);
200 201
		if (IS_ERR(policy->v.zonelist)) {
			void *error_code = policy->v.zonelist;
L
Linus Torvalds 已提交
202
			kmem_cache_free(policy_cache, policy);
203
			return error_code;
L
Linus Torvalds 已提交
204 205 206 207
		}
		break;
	}
	policy->policy = mode;
208
	policy->cpuset_mems_allowed = cpuset_mems_allowed(current);
L
Linus Torvalds 已提交
209 210 211
	return policy;
}

C
Christoph Lameter 已提交
212
static void gather_stats(struct page *, void *, int pte_dirty);
213 214
static void migrate_page_add(struct page *page, struct list_head *pagelist,
				unsigned long flags);
215

216
/* Scan through pages checking if pages follow certain conditions. */
N
Nick Piggin 已提交
217
static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
218 219
		unsigned long addr, unsigned long end,
		const nodemask_t *nodes, unsigned long flags,
220
		void *private)
L
Linus Torvalds 已提交
221
{
222 223
	pte_t *orig_pte;
	pte_t *pte;
224
	spinlock_t *ptl;
225

226
	orig_pte = pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
227
	do {
228
		struct page *page;
229
		int nid;
230 231

		if (!pte_present(*pte))
L
Linus Torvalds 已提交
232
			continue;
233 234
		page = vm_normal_page(vma, addr, *pte);
		if (!page)
L
Linus Torvalds 已提交
235
			continue;
236 237 238 239 240 241 242 243 244 245 246
		/*
		 * The check for PageReserved here is important to avoid
		 * handling zero pages and other pages that may have been
		 * marked special by the system.
		 *
		 * If the PageReserved would not be checked here then f.e.
		 * the location of the zero page could have an influence
		 * on MPOL_MF_STRICT, zero pages would be counted for
		 * the per node stats, and there would be useless attempts
		 * to put zero pages on the migration list.
		 */
247 248
		if (PageReserved(page))
			continue;
249
		nid = page_to_nid(page);
250 251 252
		if (node_isset(nid, *nodes) == !!(flags & MPOL_MF_INVERT))
			continue;

253
		if (flags & MPOL_MF_STATS)
C
Christoph Lameter 已提交
254
			gather_stats(page, private, pte_dirty(*pte));
255
		else if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
256
			migrate_page_add(page, private, flags);
257 258
		else
			break;
259
	} while (pte++, addr += PAGE_SIZE, addr != end);
260
	pte_unmap_unlock(orig_pte, ptl);
261 262 263
	return addr != end;
}

N
Nick Piggin 已提交
264
static inline int check_pmd_range(struct vm_area_struct *vma, pud_t *pud,
265 266
		unsigned long addr, unsigned long end,
		const nodemask_t *nodes, unsigned long flags,
267
		void *private)
268 269 270 271 272 273 274 275 276
{
	pmd_t *pmd;
	unsigned long next;

	pmd = pmd_offset(pud, addr);
	do {
		next = pmd_addr_end(addr, end);
		if (pmd_none_or_clear_bad(pmd))
			continue;
277
		if (check_pte_range(vma, pmd, addr, next, nodes,
278
				    flags, private))
279 280 281 282 283
			return -EIO;
	} while (pmd++, addr = next, addr != end);
	return 0;
}

N
Nick Piggin 已提交
284
static inline int check_pud_range(struct vm_area_struct *vma, pgd_t *pgd,
285 286
		unsigned long addr, unsigned long end,
		const nodemask_t *nodes, unsigned long flags,
287
		void *private)
288 289 290 291 292 293 294 295 296
{
	pud_t *pud;
	unsigned long next;

	pud = pud_offset(pgd, addr);
	do {
		next = pud_addr_end(addr, end);
		if (pud_none_or_clear_bad(pud))
			continue;
297
		if (check_pmd_range(vma, pud, addr, next, nodes,
298
				    flags, private))
299 300 301 302 303
			return -EIO;
	} while (pud++, addr = next, addr != end);
	return 0;
}

N
Nick Piggin 已提交
304
static inline int check_pgd_range(struct vm_area_struct *vma,
305 306
		unsigned long addr, unsigned long end,
		const nodemask_t *nodes, unsigned long flags,
307
		void *private)
308 309 310 311
{
	pgd_t *pgd;
	unsigned long next;

N
Nick Piggin 已提交
312
	pgd = pgd_offset(vma->vm_mm, addr);
313 314 315 316
	do {
		next = pgd_addr_end(addr, end);
		if (pgd_none_or_clear_bad(pgd))
			continue;
317
		if (check_pud_range(vma, pgd, addr, next, nodes,
318
				    flags, private))
319 320 321
			return -EIO;
	} while (pgd++, addr = next, addr != end);
	return 0;
L
Linus Torvalds 已提交
322 323
}

324 325 326 327 328
/*
 * Check if all pages in a range are on a set of nodes.
 * If pagelist != NULL then isolate pages from the LRU and
 * put them on the pagelist.
 */
L
Linus Torvalds 已提交
329 330
static struct vm_area_struct *
check_range(struct mm_struct *mm, unsigned long start, unsigned long end,
331
		const nodemask_t *nodes, unsigned long flags, void *private)
L
Linus Torvalds 已提交
332 333 334 335
{
	int err;
	struct vm_area_struct *first, *vma, *prev;

336 337
	if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) {

C
Christoph Lameter 已提交
338 339 340
		err = migrate_prep();
		if (err)
			return ERR_PTR(err);
341
	}
342

L
Linus Torvalds 已提交
343 344 345 346 347
	first = find_vma(mm, start);
	if (!first)
		return ERR_PTR(-EFAULT);
	prev = NULL;
	for (vma = first; vma && vma->vm_start < end; vma = vma->vm_next) {
348 349 350 351 352 353 354 355 356 357
		if (!(flags & MPOL_MF_DISCONTIG_OK)) {
			if (!vma->vm_next && vma->vm_end < end)
				return ERR_PTR(-EFAULT);
			if (prev && prev->vm_end < vma->vm_start)
				return ERR_PTR(-EFAULT);
		}
		if (!is_vm_hugetlb_page(vma) &&
		    ((flags & MPOL_MF_STRICT) ||
		     ((flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) &&
				vma_migratable(vma)))) {
A
Andi Kleen 已提交
358
			unsigned long endvma = vma->vm_end;
359

A
Andi Kleen 已提交
360 361 362 363
			if (endvma > end)
				endvma = end;
			if (vma->vm_start > start)
				start = vma->vm_start;
364
			err = check_pgd_range(vma, start, endvma, nodes,
365
						flags, private);
L
Linus Torvalds 已提交
366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381
			if (err) {
				first = ERR_PTR(err);
				break;
			}
		}
		prev = vma;
	}
	return first;
}

/* Apply policy to a single VMA */
static int policy_vma(struct vm_area_struct *vma, struct mempolicy *new)
{
	int err = 0;
	struct mempolicy *old = vma->vm_policy;

382
	pr_debug("vma %lx-%lx/%lx vm_ops %p vm_file %p set_policy %p\n",
L
Linus Torvalds 已提交
383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418
		 vma->vm_start, vma->vm_end, vma->vm_pgoff,
		 vma->vm_ops, vma->vm_file,
		 vma->vm_ops ? vma->vm_ops->set_policy : NULL);

	if (vma->vm_ops && vma->vm_ops->set_policy)
		err = vma->vm_ops->set_policy(vma, new);
	if (!err) {
		mpol_get(new);
		vma->vm_policy = new;
		mpol_free(old);
	}
	return err;
}

/* Step 2: apply policy to a range and do splits. */
static int mbind_range(struct vm_area_struct *vma, unsigned long start,
		       unsigned long end, struct mempolicy *new)
{
	struct vm_area_struct *next;
	int err;

	err = 0;
	for (; vma && vma->vm_start < end; vma = next) {
		next = vma->vm_next;
		if (vma->vm_start < start)
			err = split_vma(vma->vm_mm, vma, start, 1);
		if (!err && vma->vm_end > end)
			err = split_vma(vma->vm_mm, vma, end, 0);
		if (!err)
			err = policy_vma(vma, new);
		if (err)
			break;
	}
	return err;
}

419 420 421 422 423
static int contextualize_policy(int mode, nodemask_t *nodes)
{
	if (!nodes)
		return 0;

424
	cpuset_update_task_memory_state();
425 426
	if (!cpuset_nodes_subset_current_mems_allowed(*nodes))
		return -EINVAL;
427 428 429
	return mpol_check_policy(mode, nodes);
}

430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460

/*
 * Update task->flags PF_MEMPOLICY bit: set iff non-default
 * mempolicy.  Allows more rapid checking of this (combined perhaps
 * with other PF_* flag bits) on memory allocation hot code paths.
 *
 * If called from outside this file, the task 'p' should -only- be
 * a newly forked child not yet visible on the task list, because
 * manipulating the task flags of a visible task is not safe.
 *
 * The above limitation is why this routine has the funny name
 * mpol_fix_fork_child_flag().
 *
 * It is also safe to call this with a task pointer of current,
 * which the static wrapper mpol_set_task_struct_flag() does,
 * for use within this file.
 */

void mpol_fix_fork_child_flag(struct task_struct *p)
{
	if (p->mempolicy)
		p->flags |= PF_MEMPOLICY;
	else
		p->flags &= ~PF_MEMPOLICY;
}

static void mpol_set_task_struct_flag(void)
{
	mpol_fix_fork_child_flag(current);
}

L
Linus Torvalds 已提交
461
/* Set the process memory policy */
462
long do_set_mempolicy(int mode, nodemask_t *nodes)
L
Linus Torvalds 已提交
463 464 465
{
	struct mempolicy *new;

466
	if (contextualize_policy(mode, nodes))
L
Linus Torvalds 已提交
467
		return -EINVAL;
468
	new = mpol_new(mode, nodes);
L
Linus Torvalds 已提交
469 470 471 472
	if (IS_ERR(new))
		return PTR_ERR(new);
	mpol_free(current->mempolicy);
	current->mempolicy = new;
473
	mpol_set_task_struct_flag();
L
Linus Torvalds 已提交
474
	if (new && new->policy == MPOL_INTERLEAVE)
475
		current->il_next = first_node(new->v.nodes);
L
Linus Torvalds 已提交
476 477 478 479
	return 0;
}

/* Fill a zone bitmap for a policy */
480
static void get_zonemask(struct mempolicy *p, nodemask_t *nodes)
L
Linus Torvalds 已提交
481 482 483
{
	int i;

484
	nodes_clear(*nodes);
L
Linus Torvalds 已提交
485 486 487
	switch (p->policy) {
	case MPOL_BIND:
		for (i = 0; p->v.zonelist->zones[i]; i++)
488
			node_set(zone_to_nid(p->v.zonelist->zones[i]),
489
				*nodes);
L
Linus Torvalds 已提交
490 491 492 493
		break;
	case MPOL_DEFAULT:
		break;
	case MPOL_INTERLEAVE:
494
		*nodes = p->v.nodes;
L
Linus Torvalds 已提交
495 496
		break;
	case MPOL_PREFERRED:
497
		/* or use current node instead of memory_map? */
L
Linus Torvalds 已提交
498
		if (p->v.preferred_node < 0)
499
			*nodes = node_states[N_HIGH_MEMORY];
L
Linus Torvalds 已提交
500
		else
501
			node_set(p->v.preferred_node, *nodes);
L
Linus Torvalds 已提交
502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521
		break;
	default:
		BUG();
	}
}

static int lookup_node(struct mm_struct *mm, unsigned long addr)
{
	struct page *p;
	int err;

	err = get_user_pages(current, mm, addr & PAGE_MASK, 1, 0, 0, &p, NULL);
	if (err >= 0) {
		err = page_to_nid(p);
		put_page(p);
	}
	return err;
}

/* Retrieve NUMA policy */
522 523
long do_get_mempolicy(int *policy, nodemask_t *nmask,
			unsigned long addr, unsigned long flags)
L
Linus Torvalds 已提交
524
{
525
	int err;
L
Linus Torvalds 已提交
526 527 528 529
	struct mm_struct *mm = current->mm;
	struct vm_area_struct *vma = NULL;
	struct mempolicy *pol = current->mempolicy;

530
	cpuset_update_task_memory_state();
531 532
	if (flags &
		~(unsigned long)(MPOL_F_NODE|MPOL_F_ADDR|MPOL_F_MEMS_ALLOWED))
L
Linus Torvalds 已提交
533
		return -EINVAL;
534 535 536 537 538 539 540 541 542

	if (flags & MPOL_F_MEMS_ALLOWED) {
		if (flags & (MPOL_F_NODE|MPOL_F_ADDR))
			return -EINVAL;
		*policy = 0;	/* just so it's initialized */
		*nmask  = cpuset_current_mems_allowed;
		return 0;
	}

L
Linus Torvalds 已提交
543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
	if (flags & MPOL_F_ADDR) {
		down_read(&mm->mmap_sem);
		vma = find_vma_intersection(mm, addr, addr+1);
		if (!vma) {
			up_read(&mm->mmap_sem);
			return -EFAULT;
		}
		if (vma->vm_ops && vma->vm_ops->get_policy)
			pol = vma->vm_ops->get_policy(vma, addr);
		else
			pol = vma->vm_policy;
	} else if (addr)
		return -EINVAL;

	if (!pol)
		pol = &default_policy;

	if (flags & MPOL_F_NODE) {
		if (flags & MPOL_F_ADDR) {
			err = lookup_node(mm, addr);
			if (err < 0)
				goto out;
565
			*policy = err;
L
Linus Torvalds 已提交
566 567
		} else if (pol == current->mempolicy &&
				pol->policy == MPOL_INTERLEAVE) {
568
			*policy = current->il_next;
L
Linus Torvalds 已提交
569 570 571 572 573
		} else {
			err = -EINVAL;
			goto out;
		}
	} else
574
		*policy = pol->policy;
L
Linus Torvalds 已提交
575 576 577 578 579 580 581

	if (vma) {
		up_read(&current->mm->mmap_sem);
		vma = NULL;
	}

	err = 0;
582 583
	if (nmask)
		get_zonemask(pol, nmask);
L
Linus Torvalds 已提交
584 585 586 587 588 589 590

 out:
	if (vma)
		up_read(&current->mm->mmap_sem);
	return err;
}

C
Christoph Lameter 已提交
591
#ifdef CONFIG_MIGRATION
592 593 594
/*
 * page migration
 */
595 596
static void migrate_page_add(struct page *page, struct list_head *pagelist,
				unsigned long flags)
597 598
{
	/*
599
	 * Avoid migrating a page that is shared with others.
600
	 */
C
Christoph Lameter 已提交
601 602
	if ((flags & MPOL_MF_MOVE_ALL) || page_mapcount(page) == 1)
		isolate_lru_page(page, pagelist);
603
}
604

605
static struct page *new_node_page(struct page *page, unsigned long node, int **x)
606
{
607
	return alloc_pages_node(node, GFP_HIGHUSER_MOVABLE, 0);
608 609
}

610 611 612 613 614 615 616 617 618 619 620 621
/*
 * Migrate pages from one node to a target node.
 * Returns error or the number of pages not migrated.
 */
int migrate_to_node(struct mm_struct *mm, int source, int dest, int flags)
{
	nodemask_t nmask;
	LIST_HEAD(pagelist);
	int err = 0;

	nodes_clear(nmask);
	node_set(source, nmask);
622

623 624 625
	check_range(mm, mm->mmap->vm_start, TASK_SIZE, &nmask,
			flags | MPOL_MF_DISCONTIG_OK, &pagelist);

626
	if (!list_empty(&pagelist))
627 628
		err = migrate_pages(&pagelist, new_node_page, dest);

629
	return err;
630 631
}

632
/*
633 634
 * Move pages between the two nodesets so as to preserve the physical
 * layout as much as possible.
635 636 637 638 639 640 641
 *
 * Returns the number of page that could not be moved.
 */
int do_migrate_pages(struct mm_struct *mm,
	const nodemask_t *from_nodes, const nodemask_t *to_nodes, int flags)
{
	LIST_HEAD(pagelist);
642 643 644
	int busy = 0;
	int err = 0;
	nodemask_t tmp;
645

646
  	down_read(&mm->mmap_sem);
647

648 649 650 651
	err = migrate_vmas(mm, from_nodes, to_nodes, flags);
	if (err)
		goto out;

652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681
/*
 * Find a 'source' bit set in 'tmp' whose corresponding 'dest'
 * bit in 'to' is not also set in 'tmp'.  Clear the found 'source'
 * bit in 'tmp', and return that <source, dest> pair for migration.
 * The pair of nodemasks 'to' and 'from' define the map.
 *
 * If no pair of bits is found that way, fallback to picking some
 * pair of 'source' and 'dest' bits that are not the same.  If the
 * 'source' and 'dest' bits are the same, this represents a node
 * that will be migrating to itself, so no pages need move.
 *
 * If no bits are left in 'tmp', or if all remaining bits left
 * in 'tmp' correspond to the same bit in 'to', return false
 * (nothing left to migrate).
 *
 * This lets us pick a pair of nodes to migrate between, such that
 * if possible the dest node is not already occupied by some other
 * source node, minimizing the risk of overloading the memory on a
 * node that would happen if we migrated incoming memory to a node
 * before migrating outgoing memory source that same node.
 *
 * A single scan of tmp is sufficient.  As we go, we remember the
 * most recent <s, d> pair that moved (s != d).  If we find a pair
 * that not only moved, but what's better, moved to an empty slot
 * (d is not set in tmp), then we break out then, with that pair.
 * Otherwise when we finish scannng from_tmp, we at least have the
 * most recent <s, d> pair that moved.  If we get all the way through
 * the scan of tmp without finding any node that moved, much less
 * moved to an empty node, then there is nothing left worth migrating.
 */
682

683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709
	tmp = *from_nodes;
	while (!nodes_empty(tmp)) {
		int s,d;
		int source = -1;
		int dest = 0;

		for_each_node_mask(s, tmp) {
			d = node_remap(s, *from_nodes, *to_nodes);
			if (s == d)
				continue;

			source = s;	/* Node moved. Memorize */
			dest = d;

			/* dest not in remaining from nodes? */
			if (!node_isset(dest, tmp))
				break;
		}
		if (source == -1)
			break;

		node_clear(source, tmp);
		err = migrate_to_node(mm, source, dest, flags);
		if (err > 0)
			busy += err;
		if (err < 0)
			break;
710
	}
711
out:
712
	up_read(&mm->mmap_sem);
713 714 715
	if (err < 0)
		return err;
	return busy;
C
Christoph Lameter 已提交
716 717 718

}

719
static struct page *new_vma_page(struct page *page, unsigned long private, int **x)
720 721 722
{
	struct vm_area_struct *vma = (struct vm_area_struct *)private;

723 724
	return alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma,
					page_address_in_vma(page, vma));
725
}
C
Christoph Lameter 已提交
726 727 728 729 730
#else

static void migrate_page_add(struct page *page, struct list_head *pagelist,
				unsigned long flags)
{
731 732
}

C
Christoph Lameter 已提交
733 734 735 736 737
int do_migrate_pages(struct mm_struct *mm,
	const nodemask_t *from_nodes, const nodemask_t *to_nodes, int flags)
{
	return -ENOSYS;
}
738

739
static struct page *new_vma_page(struct page *page, unsigned long private, int **x)
740 741 742
{
	return NULL;
}
C
Christoph Lameter 已提交
743 744
#endif

745 746 747 748 749 750 751 752 753 754 755 756 757 758
long do_mbind(unsigned long start, unsigned long len,
		unsigned long mode, nodemask_t *nmask, unsigned long flags)
{
	struct vm_area_struct *vma;
	struct mm_struct *mm = current->mm;
	struct mempolicy *new;
	unsigned long end;
	int err;
	LIST_HEAD(pagelist);

	if ((flags & ~(unsigned long)(MPOL_MF_STRICT |
				      MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
	    || mode > MPOL_MAX)
		return -EINVAL;
759
	if ((flags & MPOL_MF_MOVE_ALL) && !capable(CAP_SYS_NICE))
760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
		return -EPERM;

	if (start & ~PAGE_MASK)
		return -EINVAL;

	if (mode == MPOL_DEFAULT)
		flags &= ~MPOL_MF_STRICT;

	len = (len + PAGE_SIZE - 1) & PAGE_MASK;
	end = start + len;

	if (end < start)
		return -EINVAL;
	if (end == start)
		return 0;

	if (mpol_check_policy(mode, nmask))
		return -EINVAL;

	new = mpol_new(mode, nmask);
	if (IS_ERR(new))
		return PTR_ERR(new);

	/*
	 * If we are using the default policy then operation
	 * on discontinuous address spaces is okay after all
	 */
	if (!new)
		flags |= MPOL_MF_DISCONTIG_OK;

790 791
	pr_debug("mbind %lx-%lx mode:%ld nodes:%lx\n",start,start+len,
		 mode, nmask ? nodes_addr(*nmask)[0] : -1);
792 793 794 795 796 797 798 799 800 801

	down_write(&mm->mmap_sem);
	vma = check_range(mm, start, end, nmask,
			  flags | MPOL_MF_INVERT, &pagelist);

	err = PTR_ERR(vma);
	if (!IS_ERR(vma)) {
		int nr_failed = 0;

		err = mbind_range(vma, start, end, new);
802

803
		if (!list_empty(&pagelist))
804 805
			nr_failed = migrate_pages(&pagelist, new_vma_page,
						(unsigned long)vma);
806 807 808 809

		if (!err && nr_failed && (flags & MPOL_MF_STRICT))
			err = -EIO;
	}
C
Christoph Lameter 已提交
810

811 812 813 814 815
	up_write(&mm->mmap_sem);
	mpol_free(new);
	return err;
}

816 817 818 819 820
/*
 * User space interface with variable sized bitmaps for nodelists.
 */

/* Copy a node mask from user space. */
821
static int get_nodes(nodemask_t *nodes, const unsigned long __user *nmask,
822 823 824 825 826 827 828 829 830 831
		     unsigned long maxnode)
{
	unsigned long k;
	unsigned long nlongs;
	unsigned long endmask;

	--maxnode;
	nodes_clear(*nodes);
	if (maxnode == 0 || !nmask)
		return 0;
A
Andi Kleen 已提交
832
	if (maxnode > PAGE_SIZE*BITS_PER_BYTE)
C
Chris Wright 已提交
833
		return -EINVAL;
834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893

	nlongs = BITS_TO_LONGS(maxnode);
	if ((maxnode % BITS_PER_LONG) == 0)
		endmask = ~0UL;
	else
		endmask = (1UL << (maxnode % BITS_PER_LONG)) - 1;

	/* When the user specified more nodes than supported just check
	   if the non supported part is all zero. */
	if (nlongs > BITS_TO_LONGS(MAX_NUMNODES)) {
		if (nlongs > PAGE_SIZE/sizeof(long))
			return -EINVAL;
		for (k = BITS_TO_LONGS(MAX_NUMNODES); k < nlongs; k++) {
			unsigned long t;
			if (get_user(t, nmask + k))
				return -EFAULT;
			if (k == nlongs - 1) {
				if (t & endmask)
					return -EINVAL;
			} else if (t)
				return -EINVAL;
		}
		nlongs = BITS_TO_LONGS(MAX_NUMNODES);
		endmask = ~0UL;
	}

	if (copy_from_user(nodes_addr(*nodes), nmask, nlongs*sizeof(unsigned long)))
		return -EFAULT;
	nodes_addr(*nodes)[nlongs-1] &= endmask;
	return 0;
}

/* Copy a kernel node mask to user space */
static int copy_nodes_to_user(unsigned long __user *mask, unsigned long maxnode,
			      nodemask_t *nodes)
{
	unsigned long copy = ALIGN(maxnode-1, 64) / 8;
	const int nbytes = BITS_TO_LONGS(MAX_NUMNODES) * sizeof(long);

	if (copy > nbytes) {
		if (copy > PAGE_SIZE)
			return -EINVAL;
		if (clear_user((char __user *)mask + nbytes, copy - nbytes))
			return -EFAULT;
		copy = nbytes;
	}
	return copy_to_user(mask, nodes_addr(*nodes), copy) ? -EFAULT : 0;
}

asmlinkage long sys_mbind(unsigned long start, unsigned long len,
			unsigned long mode,
			unsigned long __user *nmask, unsigned long maxnode,
			unsigned flags)
{
	nodemask_t nodes;
	int err;

	err = get_nodes(&nodes, nmask, maxnode);
	if (err)
		return err;
894 895 896 897
#ifdef CONFIG_CPUSETS
	/* Restrict the nodes to the allowed nodes in the cpuset */
	nodes_and(nodes, nodes, current->mems_allowed);
#endif
898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915
	return do_mbind(start, len, mode, &nodes, flags);
}

/* Set the process memory policy */
asmlinkage long sys_set_mempolicy(int mode, unsigned long __user *nmask,
		unsigned long maxnode)
{
	int err;
	nodemask_t nodes;

	if (mode < 0 || mode > MPOL_MAX)
		return -EINVAL;
	err = get_nodes(&nodes, nmask, maxnode);
	if (err)
		return err;
	return do_set_mempolicy(mode, &nodes);
}

916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950
asmlinkage long sys_migrate_pages(pid_t pid, unsigned long maxnode,
		const unsigned long __user *old_nodes,
		const unsigned long __user *new_nodes)
{
	struct mm_struct *mm;
	struct task_struct *task;
	nodemask_t old;
	nodemask_t new;
	nodemask_t task_nodes;
	int err;

	err = get_nodes(&old, old_nodes, maxnode);
	if (err)
		return err;

	err = get_nodes(&new, new_nodes, maxnode);
	if (err)
		return err;

	/* Find the mm_struct */
	read_lock(&tasklist_lock);
	task = pid ? find_task_by_pid(pid) : current;
	if (!task) {
		read_unlock(&tasklist_lock);
		return -ESRCH;
	}
	mm = get_task_mm(task);
	read_unlock(&tasklist_lock);

	if (!mm)
		return -EINVAL;

	/*
	 * Check if this process has the right to modify the specified
	 * process. The right exists if the process has administrative
A
Alexey Dobriyan 已提交
951
	 * capabilities, superuser privileges or the same
952 953 954 955
	 * userid as the target process.
	 */
	if ((current->euid != task->suid) && (current->euid != task->uid) &&
	    (current->uid != task->suid) && (current->uid != task->uid) &&
956
	    !capable(CAP_SYS_NICE)) {
957 958 959 960 961 962
		err = -EPERM;
		goto out;
	}

	task_nodes = cpuset_mems_allowed(task);
	/* Is the user allowed to access the target nodes? */
963
	if (!nodes_subset(new, task_nodes) && !capable(CAP_SYS_NICE)) {
964 965 966 967
		err = -EPERM;
		goto out;
	}

968
	if (!nodes_subset(new, node_states[N_HIGH_MEMORY])) {
969 970 971 972
		err = -EINVAL;
		goto out;
	}

973 974 975 976
	err = security_task_movememory(task);
	if (err)
		goto out;

977
	err = do_migrate_pages(mm, &old, &new,
978
		capable(CAP_SYS_NICE) ? MPOL_MF_MOVE_ALL : MPOL_MF_MOVE);
979 980 981 982 983 984
out:
	mmput(mm);
	return err;
}


985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
/* Retrieve NUMA policy */
asmlinkage long sys_get_mempolicy(int __user *policy,
				unsigned long __user *nmask,
				unsigned long maxnode,
				unsigned long addr, unsigned long flags)
{
	int err, pval;
	nodemask_t nodes;

	if (nmask != NULL && maxnode < MAX_NUMNODES)
		return -EINVAL;

	err = do_get_mempolicy(&pval, &nodes, addr, flags);

	if (err)
		return err;

	if (policy && put_user(pval, policy))
		return -EFAULT;

	if (nmask)
		err = copy_nodes_to_user(nmask, maxnode, &nodes);

	return err;
}

L
Linus Torvalds 已提交
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070
#ifdef CONFIG_COMPAT

asmlinkage long compat_sys_get_mempolicy(int __user *policy,
				     compat_ulong_t __user *nmask,
				     compat_ulong_t maxnode,
				     compat_ulong_t addr, compat_ulong_t flags)
{
	long err;
	unsigned long __user *nm = NULL;
	unsigned long nr_bits, alloc_size;
	DECLARE_BITMAP(bm, MAX_NUMNODES);

	nr_bits = min_t(unsigned long, maxnode-1, MAX_NUMNODES);
	alloc_size = ALIGN(nr_bits, BITS_PER_LONG) / 8;

	if (nmask)
		nm = compat_alloc_user_space(alloc_size);

	err = sys_get_mempolicy(policy, nm, nr_bits+1, addr, flags);

	if (!err && nmask) {
		err = copy_from_user(bm, nm, alloc_size);
		/* ensure entire bitmap is zeroed */
		err |= clear_user(nmask, ALIGN(maxnode-1, 8) / 8);
		err |= compat_put_bitmap(nmask, bm, nr_bits);
	}

	return err;
}

asmlinkage long compat_sys_set_mempolicy(int mode, compat_ulong_t __user *nmask,
				     compat_ulong_t maxnode)
{
	long err = 0;
	unsigned long __user *nm = NULL;
	unsigned long nr_bits, alloc_size;
	DECLARE_BITMAP(bm, MAX_NUMNODES);

	nr_bits = min_t(unsigned long, maxnode-1, MAX_NUMNODES);
	alloc_size = ALIGN(nr_bits, BITS_PER_LONG) / 8;

	if (nmask) {
		err = compat_get_bitmap(bm, nmask, nr_bits);
		nm = compat_alloc_user_space(alloc_size);
		err |= copy_to_user(nm, bm, alloc_size);
	}

	if (err)
		return -EFAULT;

	return sys_set_mempolicy(mode, nm, nr_bits+1);
}

asmlinkage long compat_sys_mbind(compat_ulong_t start, compat_ulong_t len,
			     compat_ulong_t mode, compat_ulong_t __user *nmask,
			     compat_ulong_t maxnode, compat_ulong_t flags)
{
	long err = 0;
	unsigned long __user *nm = NULL;
	unsigned long nr_bits, alloc_size;
1071
	nodemask_t bm;
L
Linus Torvalds 已提交
1072 1073 1074 1075 1076

	nr_bits = min_t(unsigned long, maxnode-1, MAX_NUMNODES);
	alloc_size = ALIGN(nr_bits, BITS_PER_LONG) / 8;

	if (nmask) {
1077
		err = compat_get_bitmap(nodes_addr(bm), nmask, nr_bits);
L
Linus Torvalds 已提交
1078
		nm = compat_alloc_user_space(alloc_size);
1079
		err |= copy_to_user(nm, nodes_addr(bm), alloc_size);
L
Linus Torvalds 已提交
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089
	}

	if (err)
		return -EFAULT;

	return sys_mbind(start, len, mode, nm, nr_bits+1, flags);
}

#endif

1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102
/*
 * get_vma_policy(@task, @vma, @addr)
 * @task - task for fallback if vma policy == default
 * @vma   - virtual memory area whose policy is sought
 * @addr  - address in @vma for shared policy lookup
 *
 * Returns effective policy for a VMA at specified address.
 * Falls back to @task or system default policy, as necessary.
 * Returned policy has extra reference count if shared, vma,
 * or some other task's policy [show_numa_maps() can pass
 * @task != current].  It is the caller's responsibility to
 * free the reference in these cases.
 */
1103 1104
static struct mempolicy * get_vma_policy(struct task_struct *task,
		struct vm_area_struct *vma, unsigned long addr)
L
Linus Torvalds 已提交
1105
{
1106
	struct mempolicy *pol = task->mempolicy;
1107
	int shared_pol = 0;
L
Linus Torvalds 已提交
1108 1109

	if (vma) {
1110
		if (vma->vm_ops && vma->vm_ops->get_policy) {
1111
			pol = vma->vm_ops->get_policy(vma, addr);
1112 1113
			shared_pol = 1;	/* if pol non-NULL, add ref below */
		} else if (vma->vm_policy &&
L
Linus Torvalds 已提交
1114 1115 1116 1117 1118
				vma->vm_policy->policy != MPOL_DEFAULT)
			pol = vma->vm_policy;
	}
	if (!pol)
		pol = &default_policy;
1119 1120
	else if (!shared_pol && pol != current->mempolicy)
		mpol_get(pol);	/* vma or other task's policy */
L
Linus Torvalds 已提交
1121 1122 1123 1124
	return pol;
}

/* Return a zonelist representing a mempolicy */
A
Al Viro 已提交
1125
static struct zonelist *zonelist_policy(gfp_t gfp, struct mempolicy *policy)
L
Linus Torvalds 已提交
1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137
{
	int nd;

	switch (policy->policy) {
	case MPOL_PREFERRED:
		nd = policy->v.preferred_node;
		if (nd < 0)
			nd = numa_node_id();
		break;
	case MPOL_BIND:
		/* Lower zones don't get a policy applied */
		/* Careful: current->mems_allowed might have moved */
1138
		if (gfp_zone(gfp) >= policy_zone)
L
Linus Torvalds 已提交
1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149
			if (cpuset_zonelist_valid_mems_allowed(policy->v.zonelist))
				return policy->v.zonelist;
		/*FALL THROUGH*/
	case MPOL_INTERLEAVE: /* should not happen */
	case MPOL_DEFAULT:
		nd = numa_node_id();
		break;
	default:
		nd = 0;
		BUG();
	}
A
Al Viro 已提交
1150
	return NODE_DATA(nd)->node_zonelists + gfp_zone(gfp);
L
Linus Torvalds 已提交
1151 1152 1153 1154 1155 1156 1157 1158 1159
}

/* Do dynamic interleaving for a process */
static unsigned interleave_nodes(struct mempolicy *policy)
{
	unsigned nid, next;
	struct task_struct *me = current;

	nid = me->il_next;
1160
	next = next_node(nid, policy->v.nodes);
L
Linus Torvalds 已提交
1161
	if (next >= MAX_NUMNODES)
1162
		next = first_node(policy->v.nodes);
L
Linus Torvalds 已提交
1163 1164 1165 1166
	me->il_next = next;
	return nid;
}

1167 1168 1169 1170 1171 1172
/*
 * Depending on the memory policy provide a node from which to allocate the
 * next slab entry.
 */
unsigned slab_node(struct mempolicy *policy)
{
1173 1174 1175
	int pol = policy ? policy->policy : MPOL_DEFAULT;

	switch (pol) {
1176 1177 1178 1179 1180 1181 1182 1183
	case MPOL_INTERLEAVE:
		return interleave_nodes(policy);

	case MPOL_BIND:
		/*
		 * Follow bind policy behavior and start allocation at the
		 * first node.
		 */
1184
		return zone_to_nid(policy->v.zonelist->zones[0]);
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195

	case MPOL_PREFERRED:
		if (policy->v.preferred_node >= 0)
			return policy->v.preferred_node;
		/* Fall through */

	default:
		return numa_node_id();
	}
}

L
Linus Torvalds 已提交
1196 1197 1198 1199
/* Do static interleaving for a VMA with known offset. */
static unsigned offset_il_node(struct mempolicy *pol,
		struct vm_area_struct *vma, unsigned long off)
{
1200
	unsigned nnodes = nodes_weight(pol->v.nodes);
L
Linus Torvalds 已提交
1201 1202 1203 1204 1205 1206
	unsigned target = (unsigned)off % nnodes;
	int c;
	int nid = -1;

	c = 0;
	do {
1207
		nid = next_node(nid, pol->v.nodes);
L
Linus Torvalds 已提交
1208 1209 1210 1211 1212
		c++;
	} while (c <= target);
	return nid;
}

1213 1214 1215 1216 1217 1218 1219
/* Determine a node number for interleave */
static inline unsigned interleave_nid(struct mempolicy *pol,
		 struct vm_area_struct *vma, unsigned long addr, int shift)
{
	if (vma) {
		unsigned long off;

1220 1221 1222 1223 1224 1225 1226 1227 1228
		/*
		 * for small pages, there is no difference between
		 * shift and PAGE_SHIFT, so the bit-shift is safe.
		 * for huge pages, since vm_pgoff is in units of small
		 * pages, we need to shift off the always 0 bits to get
		 * a useful offset.
		 */
		BUG_ON(shift < PAGE_SHIFT);
		off = vma->vm_pgoff >> (shift - PAGE_SHIFT);
1229 1230 1231 1232 1233 1234
		off += (addr - vma->vm_start) >> shift;
		return offset_il_node(pol, vma, off);
	} else
		return interleave_nodes(pol);
}

1235
#ifdef CONFIG_HUGETLBFS
1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251
/*
 * huge_zonelist(@vma, @addr, @gfp_flags, @mpol)
 * @vma = virtual memory area whose policy is sought
 * @addr = address in @vma for shared policy lookup and interleave policy
 * @gfp_flags = for requested zone
 * @mpol = pointer to mempolicy pointer for reference counted 'BIND policy
 *
 * Returns a zonelist suitable for a huge page allocation.
 * If the effective policy is 'BIND, returns pointer to policy's zonelist.
 * If it is also a policy for which get_vma_policy() returns an extra
 * reference, we must hold that reference until after allocation.
 * In that case, return policy via @mpol so hugetlb allocation can drop
 * the reference.  For non-'BIND referenced policies, we can/do drop the
 * reference here, so the caller doesn't need to know about the special case
 * for default and current task policy.
 */
1252
struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr,
1253
				gfp_t gfp_flags, struct mempolicy **mpol)
1254 1255
{
	struct mempolicy *pol = get_vma_policy(current, vma, addr);
1256
	struct zonelist *zl;
1257

1258
	*mpol = NULL;		/* probably no unref needed */
1259 1260 1261 1262
	if (pol->policy == MPOL_INTERLEAVE) {
		unsigned nid;

		nid = interleave_nid(pol, vma, addr, HPAGE_SHIFT);
1263
		__mpol_free(pol);		/* finished with pol */
1264
		return NODE_DATA(nid)->node_zonelists + gfp_zone(gfp_flags);
1265
	}
1266 1267 1268 1269 1270 1271 1272 1273 1274

	zl = zonelist_policy(GFP_HIGHUSER, pol);
	if (unlikely(pol != &default_policy && pol != current->mempolicy)) {
		if (pol->policy != MPOL_BIND)
			__mpol_free(pol);	/* finished with pol */
		else
			*mpol = pol;	/* unref needed after allocation */
	}
	return zl;
1275
}
1276
#endif
1277

L
Linus Torvalds 已提交
1278 1279
/* Allocate a page in interleaved policy.
   Own path because it needs to do special accounting. */
1280 1281
static struct page *alloc_page_interleave(gfp_t gfp, unsigned order,
					unsigned nid)
L
Linus Torvalds 已提交
1282 1283 1284 1285
{
	struct zonelist *zl;
	struct page *page;

A
Al Viro 已提交
1286
	zl = NODE_DATA(nid)->node_zonelists + gfp_zone(gfp);
L
Linus Torvalds 已提交
1287
	page = __alloc_pages(gfp, order, zl);
1288 1289
	if (page && page_zone(page) == zl->zones[0])
		inc_zone_page_state(page, NUMA_INTERLEAVE_HIT);
L
Linus Torvalds 已提交
1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
	return page;
}

/**
 * 	alloc_page_vma	- Allocate a page for a VMA.
 *
 * 	@gfp:
 *      %GFP_USER    user allocation.
 *      %GFP_KERNEL  kernel allocations,
 *      %GFP_HIGHMEM highmem/user allocations,
 *      %GFP_FS      allocation should not call back into a file system.
 *      %GFP_ATOMIC  don't sleep.
 *
 * 	@vma:  Pointer to VMA or NULL if not available.
 *	@addr: Virtual Address of the allocation. Must be inside the VMA.
 *
 * 	This function allocates a page from the kernel page pool and applies
 *	a NUMA policy associated with the VMA or the current process.
 *	When VMA is not NULL caller must hold down_read on the mmap_sem of the
 *	mm_struct of the VMA to prevent it from going away. Should be used for
 *	all allocations for pages that will be mapped into
 * 	user space. Returns NULL when no page can be allocated.
 *
 *	Should be called with the mm_sem of the vma hold.
 */
struct page *
A
Al Viro 已提交
1316
alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr)
L
Linus Torvalds 已提交
1317
{
1318
	struct mempolicy *pol = get_vma_policy(current, vma, addr);
1319
	struct zonelist *zl;
L
Linus Torvalds 已提交
1320

1321
	cpuset_update_task_memory_state();
L
Linus Torvalds 已提交
1322 1323 1324

	if (unlikely(pol->policy == MPOL_INTERLEAVE)) {
		unsigned nid;
1325 1326

		nid = interleave_nid(pol, vma, addr, PAGE_SHIFT);
L
Linus Torvalds 已提交
1327 1328
		return alloc_page_interleave(gfp, 0, nid);
	}
1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
	zl = zonelist_policy(gfp, pol);
	if (pol != &default_policy && pol != current->mempolicy) {
		/*
		 * slow path: ref counted policy -- shared or vma
		 */
		struct page *page =  __alloc_pages(gfp, 0, zl);
		__mpol_free(pol);
		return page;
	}
	/*
	 * fast path:  default or task policy
	 */
	return __alloc_pages(gfp, 0, zl);
L
Linus Torvalds 已提交
1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358
}

/**
 * 	alloc_pages_current - Allocate pages.
 *
 *	@gfp:
 *		%GFP_USER   user allocation,
 *      	%GFP_KERNEL kernel allocation,
 *      	%GFP_HIGHMEM highmem allocation,
 *      	%GFP_FS     don't call back into a file system.
 *      	%GFP_ATOMIC don't sleep.
 *	@order: Power of two of allocation size in pages. 0 is a single page.
 *
 *	Allocate a page from the kernel page pool.  When not in
 *	interrupt context and apply the current process NUMA policy.
 *	Returns NULL when no page can be allocated.
 *
1359
 *	Don't call cpuset_update_task_memory_state() unless
L
Linus Torvalds 已提交
1360 1361 1362
 *	1) it's ok to take cpuset_sem (can WAIT), and
 *	2) allocating for current task (not interrupt).
 */
A
Al Viro 已提交
1363
struct page *alloc_pages_current(gfp_t gfp, unsigned order)
L
Linus Torvalds 已提交
1364 1365 1366 1367
{
	struct mempolicy *pol = current->mempolicy;

	if ((gfp & __GFP_WAIT) && !in_interrupt())
1368
		cpuset_update_task_memory_state();
1369
	if (!pol || in_interrupt() || (gfp & __GFP_THISNODE))
L
Linus Torvalds 已提交
1370 1371 1372 1373 1374 1375 1376
		pol = &default_policy;
	if (pol->policy == MPOL_INTERLEAVE)
		return alloc_page_interleave(gfp, order, interleave_nodes(pol));
	return __alloc_pages(gfp, order, zonelist_policy(gfp, pol));
}
EXPORT_SYMBOL(alloc_pages_current);

1377 1378 1379 1380 1381 1382 1383 1384 1385
/*
 * If mpol_copy() sees current->cpuset == cpuset_being_rebound, then it
 * rebinds the mempolicy its copying by calling mpol_rebind_policy()
 * with the mems_allowed returned by cpuset_mems_allowed().  This
 * keeps mempolicies cpuset relative after its cpuset moves.  See
 * further kernel/cpuset.c update_nodemask().
 */
void *cpuset_being_rebound;

L
Linus Torvalds 已提交
1386 1387 1388 1389 1390 1391 1392
/* Slow path of a mempolicy copy */
struct mempolicy *__mpol_copy(struct mempolicy *old)
{
	struct mempolicy *new = kmem_cache_alloc(policy_cache, GFP_KERNEL);

	if (!new)
		return ERR_PTR(-ENOMEM);
1393 1394 1395 1396
	if (current_cpuset_is_being_rebound()) {
		nodemask_t mems = cpuset_mems_allowed(current);
		mpol_rebind_policy(old, &mems);
	}
L
Linus Torvalds 已提交
1397 1398 1399 1400
	*new = *old;
	atomic_set(&new->refcnt, 1);
	if (new->policy == MPOL_BIND) {
		int sz = ksize(old->v.zonelist);
1401
		new->v.zonelist = kmemdup(old->v.zonelist, sz, GFP_KERNEL);
L
Linus Torvalds 已提交
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420
		if (!new->v.zonelist) {
			kmem_cache_free(policy_cache, new);
			return ERR_PTR(-ENOMEM);
		}
	}
	return new;
}

/* Slow path of a mempolicy comparison */
int __mpol_equal(struct mempolicy *a, struct mempolicy *b)
{
	if (!a || !b)
		return 0;
	if (a->policy != b->policy)
		return 0;
	switch (a->policy) {
	case MPOL_DEFAULT:
		return 1;
	case MPOL_INTERLEAVE:
1421
		return nodes_equal(a->v.nodes, b->v.nodes);
L
Linus Torvalds 已提交
1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
	case MPOL_PREFERRED:
		return a->v.preferred_node == b->v.preferred_node;
	case MPOL_BIND: {
		int i;
		for (i = 0; a->v.zonelist->zones[i]; i++)
			if (a->v.zonelist->zones[i] != b->v.zonelist->zones[i])
				return 0;
		return b->v.zonelist->zones[i] == NULL;
	}
	default:
		BUG();
		return 0;
	}
}

/* Slow path of a mpol destructor. */
void __mpol_free(struct mempolicy *p)
{
	if (!atomic_dec_and_test(&p->refcnt))
		return;
	if (p->policy == MPOL_BIND)
		kfree(p->v.zonelist);
	p->policy = MPOL_DEFAULT;
	kmem_cache_free(policy_cache, p);
}

/*
 * Shared memory backing store policy support.
 *
 * Remember policies even when nobody has shared memory mapped.
 * The policies are kept in Red-Black tree linked from the inode.
 * They are protected by the sp->lock spinlock, which should be held
 * for any accesses to the tree.
 */

/* lookup first element intersecting start-end */
/* Caller holds sp->lock */
static struct sp_node *
sp_lookup(struct shared_policy *sp, unsigned long start, unsigned long end)
{
	struct rb_node *n = sp->root.rb_node;

	while (n) {
		struct sp_node *p = rb_entry(n, struct sp_node, nd);

		if (start >= p->end)
			n = n->rb_right;
		else if (end <= p->start)
			n = n->rb_left;
		else
			break;
	}
	if (!n)
		return NULL;
	for (;;) {
		struct sp_node *w = NULL;
		struct rb_node *prev = rb_prev(n);
		if (!prev)
			break;
		w = rb_entry(prev, struct sp_node, nd);
		if (w->end <= start)
			break;
		n = prev;
	}
	return rb_entry(n, struct sp_node, nd);
}

/* Insert a new shared policy into the list. */
/* Caller holds sp->lock */
static void sp_insert(struct shared_policy *sp, struct sp_node *new)
{
	struct rb_node **p = &sp->root.rb_node;
	struct rb_node *parent = NULL;
	struct sp_node *nd;

	while (*p) {
		parent = *p;
		nd = rb_entry(parent, struct sp_node, nd);
		if (new->start < nd->start)
			p = &(*p)->rb_left;
		else if (new->end > nd->end)
			p = &(*p)->rb_right;
		else
			BUG();
	}
	rb_link_node(&new->nd, parent, p);
	rb_insert_color(&new->nd, &sp->root);
1509
	pr_debug("inserting %lx-%lx: %d\n", new->start, new->end,
L
Linus Torvalds 已提交
1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533
		 new->policy ? new->policy->policy : 0);
}

/* Find shared policy intersecting idx */
struct mempolicy *
mpol_shared_policy_lookup(struct shared_policy *sp, unsigned long idx)
{
	struct mempolicy *pol = NULL;
	struct sp_node *sn;

	if (!sp->root.rb_node)
		return NULL;
	spin_lock(&sp->lock);
	sn = sp_lookup(sp, idx, idx+1);
	if (sn) {
		mpol_get(sn->policy);
		pol = sn->policy;
	}
	spin_unlock(&sp->lock);
	return pol;
}

static void sp_delete(struct shared_policy *sp, struct sp_node *n)
{
1534
	pr_debug("deleting %lx-l%lx\n", n->start, n->end);
L
Linus Torvalds 已提交
1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601
	rb_erase(&n->nd, &sp->root);
	mpol_free(n->policy);
	kmem_cache_free(sn_cache, n);
}

struct sp_node *
sp_alloc(unsigned long start, unsigned long end, struct mempolicy *pol)
{
	struct sp_node *n = kmem_cache_alloc(sn_cache, GFP_KERNEL);

	if (!n)
		return NULL;
	n->start = start;
	n->end = end;
	mpol_get(pol);
	n->policy = pol;
	return n;
}

/* Replace a policy range. */
static int shared_policy_replace(struct shared_policy *sp, unsigned long start,
				 unsigned long end, struct sp_node *new)
{
	struct sp_node *n, *new2 = NULL;

restart:
	spin_lock(&sp->lock);
	n = sp_lookup(sp, start, end);
	/* Take care of old policies in the same range. */
	while (n && n->start < end) {
		struct rb_node *next = rb_next(&n->nd);
		if (n->start >= start) {
			if (n->end <= end)
				sp_delete(sp, n);
			else
				n->start = end;
		} else {
			/* Old policy spanning whole new range. */
			if (n->end > end) {
				if (!new2) {
					spin_unlock(&sp->lock);
					new2 = sp_alloc(end, n->end, n->policy);
					if (!new2)
						return -ENOMEM;
					goto restart;
				}
				n->end = start;
				sp_insert(sp, new2);
				new2 = NULL;
				break;
			} else
				n->end = start;
		}
		if (!next)
			break;
		n = rb_entry(next, struct sp_node, nd);
	}
	if (new)
		sp_insert(sp, new);
	spin_unlock(&sp->lock);
	if (new2) {
		mpol_free(new2->policy);
		kmem_cache_free(sn_cache, new2);
	}
	return 0;
}

1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625
void mpol_shared_policy_init(struct shared_policy *info, int policy,
				nodemask_t *policy_nodes)
{
	info->root = RB_ROOT;
	spin_lock_init(&info->lock);

	if (policy != MPOL_DEFAULT) {
		struct mempolicy *newpol;

		/* Falls back to MPOL_DEFAULT on any error */
		newpol = mpol_new(policy, policy_nodes);
		if (!IS_ERR(newpol)) {
			/* Create pseudo-vma that contains just the policy */
			struct vm_area_struct pvma;

			memset(&pvma, 0, sizeof(struct vm_area_struct));
			/* Policy covers entire file */
			pvma.vm_end = TASK_SIZE;
			mpol_set_shared_policy(info, &pvma, newpol);
			mpol_free(newpol);
		}
	}
}

L
Linus Torvalds 已提交
1626 1627 1628 1629 1630 1631 1632
int mpol_set_shared_policy(struct shared_policy *info,
			struct vm_area_struct *vma, struct mempolicy *npol)
{
	int err;
	struct sp_node *new = NULL;
	unsigned long sz = vma_pages(vma);

1633
	pr_debug("set_shared_policy %lx sz %lu %d %lx\n",
L
Linus Torvalds 已提交
1634 1635
		 vma->vm_pgoff,
		 sz, npol? npol->policy : -1,
1636
		 npol ? nodes_addr(npol->v.nodes)[0] : -1);
L
Linus Torvalds 已提交
1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661

	if (npol) {
		new = sp_alloc(vma->vm_pgoff, vma->vm_pgoff + sz, npol);
		if (!new)
			return -ENOMEM;
	}
	err = shared_policy_replace(info, vma->vm_pgoff, vma->vm_pgoff+sz, new);
	if (err && new)
		kmem_cache_free(sn_cache, new);
	return err;
}

/* Free a backing policy store on inode delete. */
void mpol_free_shared_policy(struct shared_policy *p)
{
	struct sp_node *n;
	struct rb_node *next;

	if (!p->root.rb_node)
		return;
	spin_lock(&p->lock);
	next = rb_first(&p->root);
	while (next) {
		n = rb_entry(next, struct sp_node, nd);
		next = rb_next(&n->nd);
1662
		rb_erase(&n->nd, &p->root);
L
Linus Torvalds 已提交
1663 1664 1665 1666 1667 1668 1669 1670 1671
		mpol_free(n->policy);
		kmem_cache_free(sn_cache, n);
	}
	spin_unlock(&p->lock);
}

/* assumes fs == KERNEL_DS */
void __init numa_policy_init(void)
{
1672 1673 1674 1675
	nodemask_t interleave_nodes;
	unsigned long largest = 0;
	int nid, prefer = 0;

L
Linus Torvalds 已提交
1676 1677
	policy_cache = kmem_cache_create("numa_policy",
					 sizeof(struct mempolicy),
1678
					 0, SLAB_PANIC, NULL);
L
Linus Torvalds 已提交
1679 1680 1681

	sn_cache = kmem_cache_create("shared_policy_node",
				     sizeof(struct sp_node),
1682
				     0, SLAB_PANIC, NULL);
L
Linus Torvalds 已提交
1683

1684 1685 1686 1687 1688 1689
	/*
	 * Set interleaving policy for system init. Interleaving is only
	 * enabled across suitably sized nodes (default is >= 16MB), or
	 * fall back to the largest node if they're all smaller.
	 */
	nodes_clear(interleave_nodes);
1690
	for_each_node_state(nid, N_HIGH_MEMORY) {
1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706
		unsigned long total_pages = node_present_pages(nid);

		/* Preserve the largest node */
		if (largest < total_pages) {
			largest = total_pages;
			prefer = nid;
		}

		/* Interleave this node? */
		if ((total_pages << PAGE_SHIFT) >= (16 << 20))
			node_set(nid, interleave_nodes);
	}

	/* All too small, use the largest */
	if (unlikely(nodes_empty(interleave_nodes)))
		node_set(prefer, interleave_nodes);
L
Linus Torvalds 已提交
1707

1708
	if (do_set_mempolicy(MPOL_INTERLEAVE, &interleave_nodes))
L
Linus Torvalds 已提交
1709 1710 1711
		printk("numa_policy_init: interleaving failed\n");
}

1712
/* Reset policy of current process to default */
L
Linus Torvalds 已提交
1713 1714
void numa_default_policy(void)
{
1715
	do_set_mempolicy(MPOL_DEFAULT, NULL);
L
Linus Torvalds 已提交
1716
}
1717 1718

/* Migrate a policy to a different set of nodes */
1719
void mpol_rebind_policy(struct mempolicy *pol, const nodemask_t *newmask)
1720
{
1721
	nodemask_t *mpolmask;
1722 1723 1724 1725
	nodemask_t tmp;

	if (!pol)
		return;
1726 1727 1728
	mpolmask = &pol->cpuset_mems_allowed;
	if (nodes_equal(*mpolmask, *newmask))
		return;
1729 1730 1731 1732 1733

	switch (pol->policy) {
	case MPOL_DEFAULT:
		break;
	case MPOL_INTERLEAVE:
1734
		nodes_remap(tmp, pol->v.nodes, *mpolmask, *newmask);
1735
		pol->v.nodes = tmp;
1736 1737 1738
		*mpolmask = *newmask;
		current->il_next = node_remap(current->il_next,
						*mpolmask, *newmask);
1739 1740 1741
		break;
	case MPOL_PREFERRED:
		pol->v.preferred_node = node_remap(pol->v.preferred_node,
1742 1743
						*mpolmask, *newmask);
		*mpolmask = *newmask;
1744 1745 1746 1747 1748 1749 1750 1751
		break;
	case MPOL_BIND: {
		nodemask_t nodes;
		struct zone **z;
		struct zonelist *zonelist;

		nodes_clear(nodes);
		for (z = pol->v.zonelist->zones; *z; z++)
1752
			node_set(zone_to_nid(*z), nodes);
1753
		nodes_remap(tmp, nodes, *mpolmask, *newmask);
1754 1755 1756 1757 1758 1759 1760 1761 1762
		nodes = tmp;

		zonelist = bind_zonelist(&nodes);

		/* If no mem, then zonelist is NULL and we keep old zonelist.
		 * If that old zonelist has no remaining mems_allowed nodes,
		 * then zonelist_policy() will "FALL THROUGH" to MPOL_DEFAULT.
		 */

1763
		if (!IS_ERR(zonelist)) {
1764 1765 1766 1767
			/* Good - got mem - substitute new zonelist */
			kfree(pol->v.zonelist);
			pol->v.zonelist = zonelist;
		}
1768
		*mpolmask = *newmask;
1769 1770 1771 1772 1773 1774 1775 1776 1777
		break;
	}
	default:
		BUG();
		break;
	}
}

/*
1778 1779
 * Wrapper for mpol_rebind_policy() that just requires task
 * pointer, and updates task mempolicy.
1780
 */
1781 1782

void mpol_rebind_task(struct task_struct *tsk, const nodemask_t *new)
1783
{
1784
	mpol_rebind_policy(tsk->mempolicy, new);
1785
}
1786

1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802
/*
 * Rebind each vma in mm to new nodemask.
 *
 * Call holding a reference to mm.  Takes mm->mmap_sem during call.
 */

void mpol_rebind_mm(struct mm_struct *mm, nodemask_t *new)
{
	struct vm_area_struct *vma;

	down_write(&mm->mmap_sem);
	for (vma = mm->mmap; vma; vma = vma->vm_next)
		mpol_rebind_policy(vma->vm_policy, new);
	up_write(&mm->mmap_sem);
}

1803 1804 1805 1806
/*
 * Display pages allocated per node and memory policy via /proc.
 */

1807 1808
static const char * const policy_types[] =
	{ "default", "prefer", "bind", "interleave" };
1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863

/*
 * Convert a mempolicy into a string.
 * Returns the number of characters in buffer (if positive)
 * or an error (negative)
 */
static inline int mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol)
{
	char *p = buffer;
	int l;
	nodemask_t nodes;
	int mode = pol ? pol->policy : MPOL_DEFAULT;

	switch (mode) {
	case MPOL_DEFAULT:
		nodes_clear(nodes);
		break;

	case MPOL_PREFERRED:
		nodes_clear(nodes);
		node_set(pol->v.preferred_node, nodes);
		break;

	case MPOL_BIND:
		get_zonemask(pol, &nodes);
		break;

	case MPOL_INTERLEAVE:
		nodes = pol->v.nodes;
		break;

	default:
		BUG();
		return -EFAULT;
	}

	l = strlen(policy_types[mode]);
 	if (buffer + maxlen < p + l + 1)
 		return -ENOSPC;

	strcpy(p, policy_types[mode]);
	p += l;

	if (!nodes_empty(nodes)) {
		if (buffer + maxlen < p + 2)
			return -ENOSPC;
		*p++ = '=';
	 	p += nodelist_scnprintf(p, buffer + maxlen - p, nodes);
	}
	return p - buffer;
}

struct numa_maps {
	unsigned long pages;
	unsigned long anon;
C
Christoph Lameter 已提交
1864 1865
	unsigned long active;
	unsigned long writeback;
1866
	unsigned long mapcount_max;
C
Christoph Lameter 已提交
1867 1868
	unsigned long dirty;
	unsigned long swapcache;
1869 1870 1871
	unsigned long node[MAX_NUMNODES];
};

C
Christoph Lameter 已提交
1872
static void gather_stats(struct page *page, void *private, int pte_dirty)
1873 1874 1875 1876
{
	struct numa_maps *md = private;
	int count = page_mapcount(page);

C
Christoph Lameter 已提交
1877 1878 1879
	md->pages++;
	if (pte_dirty || PageDirty(page))
		md->dirty++;
1880

C
Christoph Lameter 已提交
1881 1882
	if (PageSwapCache(page))
		md->swapcache++;
1883

C
Christoph Lameter 已提交
1884 1885 1886 1887 1888
	if (PageActive(page))
		md->active++;

	if (PageWriteback(page))
		md->writeback++;
1889 1890 1891 1892

	if (PageAnon(page))
		md->anon++;

C
Christoph Lameter 已提交
1893 1894 1895
	if (count > md->mapcount_max)
		md->mapcount_max = count;

1896 1897 1898
	md->node[page_to_nid(page)]++;
}

A
Andrew Morton 已提交
1899
#ifdef CONFIG_HUGETLB_PAGE
C
Christoph Lameter 已提交
1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924
static void check_huge_range(struct vm_area_struct *vma,
		unsigned long start, unsigned long end,
		struct numa_maps *md)
{
	unsigned long addr;
	struct page *page;

	for (addr = start; addr < end; addr += HPAGE_SIZE) {
		pte_t *ptep = huge_pte_offset(vma->vm_mm, addr & HPAGE_MASK);
		pte_t pte;

		if (!ptep)
			continue;

		pte = *ptep;
		if (pte_none(pte))
			continue;

		page = pte_page(pte);
		if (!page)
			continue;

		gather_stats(page, md, pte_dirty(*ptep));
	}
}
A
Andrew Morton 已提交
1925 1926 1927 1928 1929 1930 1931
#else
static inline void check_huge_range(struct vm_area_struct *vma,
		unsigned long start, unsigned long end,
		struct numa_maps *md)
{
}
#endif
C
Christoph Lameter 已提交
1932

1933 1934
int show_numa_map(struct seq_file *m, void *v)
{
1935
	struct proc_maps_private *priv = m->private;
1936 1937
	struct vm_area_struct *vma = v;
	struct numa_maps *md;
C
Christoph Lameter 已提交
1938 1939
	struct file *file = vma->vm_file;
	struct mm_struct *mm = vma->vm_mm;
1940
	struct mempolicy *pol;
1941 1942 1943
	int n;
	char buffer[50];

C
Christoph Lameter 已提交
1944
	if (!mm)
1945 1946 1947 1948 1949 1950
		return 0;

	md = kzalloc(sizeof(struct numa_maps), GFP_KERNEL);
	if (!md)
		return 0;

1951 1952 1953 1954 1955 1956 1957
	pol = get_vma_policy(priv->task, vma, vma->vm_start);
	mpol_to_str(buffer, sizeof(buffer), pol);
	/*
	 * unref shared or other task's mempolicy
	 */
	if (pol != &default_policy && pol != current->mempolicy)
		__mpol_free(pol);
C
Christoph Lameter 已提交
1958 1959 1960 1961 1962

	seq_printf(m, "%08lx %s", vma->vm_start, buffer);

	if (file) {
		seq_printf(m, " file=");
J
Josef Sipek 已提交
1963
		seq_path(m, file->f_path.mnt, file->f_path.dentry, "\n\t= ");
C
Christoph Lameter 已提交
1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974
	} else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) {
		seq_printf(m, " heap");
	} else if (vma->vm_start <= mm->start_stack &&
			vma->vm_end >= mm->start_stack) {
		seq_printf(m, " stack");
	}

	if (is_vm_hugetlb_page(vma)) {
		check_huge_range(vma, vma->vm_start, vma->vm_end, md);
		seq_printf(m, " huge");
	} else {
1975
		check_pgd_range(vma, vma->vm_start, vma->vm_end,
1976
			&node_states[N_HIGH_MEMORY], MPOL_MF_STATS, md);
C
Christoph Lameter 已提交
1977 1978 1979 1980
	}

	if (!md->pages)
		goto out;
1981

C
Christoph Lameter 已提交
1982 1983
	if (md->anon)
		seq_printf(m," anon=%lu",md->anon);
1984

C
Christoph Lameter 已提交
1985 1986
	if (md->dirty)
		seq_printf(m," dirty=%lu",md->dirty);
1987

C
Christoph Lameter 已提交
1988 1989
	if (md->pages != md->anon && md->pages != md->dirty)
		seq_printf(m, " mapped=%lu", md->pages);
1990

C
Christoph Lameter 已提交
1991 1992
	if (md->mapcount_max > 1)
		seq_printf(m, " mapmax=%lu", md->mapcount_max);
1993

C
Christoph Lameter 已提交
1994 1995 1996 1997 1998 1999 2000 2001 2002
	if (md->swapcache)
		seq_printf(m," swapcache=%lu", md->swapcache);

	if (md->active < md->pages && !is_vm_hugetlb_page(vma))
		seq_printf(m," active=%lu", md->active);

	if (md->writeback)
		seq_printf(m," writeback=%lu", md->writeback);

2003
	for_each_node_state(n, N_HIGH_MEMORY)
C
Christoph Lameter 已提交
2004 2005 2006 2007
		if (md->node[n])
			seq_printf(m, " N%d=%lu", n, md->node[n]);
out:
	seq_putc(m, '\n');
2008 2009 2010
	kfree(md);

	if (m->count < m->size)
2011
		m->version = (vma != priv->tail_vma) ? vma->vm_start : 0;
2012 2013 2014
	return 0;
}