blob: c2b2a94f9d6773d1be1aece2387d6a6a52b1ae33 [file] [log] [blame]
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001/*
Izik Eidus31dbd012009-09-21 17:02:03 -07002 * Memory merging support.
3 *
4 * This code enables dynamic sharing of identical pages found in different
5 * memory areas, even if they are not shared by fork()
6 *
Izik Eidus36b25282009-09-21 17:02:06 -07007 * Copyright (C) 2008-2009 Red Hat, Inc.
Izik Eidus31dbd012009-09-21 17:02:03 -07008 * Authors:
9 * Izik Eidus
10 * Andrea Arcangeli
11 * Chris Wright
Izik Eidus36b25282009-09-21 17:02:06 -070012 * Hugh Dickins
Izik Eidus31dbd012009-09-21 17:02:03 -070013 *
14 * This work is licensed under the terms of the GNU GPL, version 2.
Hugh Dickinsf8af4da2009-09-21 17:01:57 -070015 */
16
17#include <linux/errno.h>
Izik Eidus31dbd012009-09-21 17:02:03 -070018#include <linux/mm.h>
19#include <linux/fs.h>
Hugh Dickinsf8af4da2009-09-21 17:01:57 -070020#include <linux/mman.h>
Izik Eidus31dbd012009-09-21 17:02:03 -070021#include <linux/sched.h>
22#include <linux/rwsem.h>
23#include <linux/pagemap.h>
24#include <linux/rmap.h>
25#include <linux/spinlock.h>
26#include <linux/jhash.h>
27#include <linux/delay.h>
28#include <linux/kthread.h>
29#include <linux/wait.h>
30#include <linux/slab.h>
31#include <linux/rbtree.h>
Hugh Dickins62b61f62009-12-14 17:59:33 -080032#include <linux/memory.h>
Izik Eidus31dbd012009-09-21 17:02:03 -070033#include <linux/mmu_notifier.h>
Izik Eidus2c6854f2009-09-23 15:56:04 -070034#include <linux/swap.h>
Hugh Dickinsf8af4da2009-09-21 17:01:57 -070035#include <linux/ksm.h>
Lai Jiangshand9f89842010-08-09 17:20:02 -070036#include <linux/hash.h>
Andrea Arcangeli878aee72011-01-13 15:47:10 -080037#include <linux/freezer.h>
Hugh Dickinsf8af4da2009-09-21 17:01:57 -070038
Izik Eidus31dbd012009-09-21 17:02:03 -070039#include <asm/tlbflush.h>
Hugh Dickins73848b42009-12-14 17:59:22 -080040#include "internal.h"
Izik Eidus31dbd012009-09-21 17:02:03 -070041
42/*
43 * A few notes about the KSM scanning process,
44 * to make it easier to understand the data structures below:
45 *
46 * In order to reduce excessive scanning, KSM sorts the memory pages by their
47 * contents into a data structure that holds pointers to the pages' locations.
48 *
49 * Since the contents of the pages may change at any moment, KSM cannot just
50 * insert the pages into a normal sorted tree and expect it to find anything.
51 * Therefore KSM uses two data structures - the stable and the unstable tree.
52 *
53 * The stable tree holds pointers to all the merged pages (ksm pages), sorted
54 * by their contents. Because each such page is write-protected, searching on
55 * this tree is fully assured to be working (except when pages are unmapped),
56 * and therefore this tree is called the stable tree.
57 *
58 * In addition to the stable tree, KSM uses a second data structure called the
59 * unstable tree: this tree holds pointers to pages which have been found to
60 * be "unchanged for a period of time". The unstable tree sorts these pages
61 * by their contents, but since they are not write-protected, KSM cannot rely
62 * upon the unstable tree to work correctly - the unstable tree is liable to
63 * be corrupted as its contents are modified, and so it is called unstable.
64 *
65 * KSM solves this problem by several techniques:
66 *
67 * 1) The unstable tree is flushed every time KSM completes scanning all
68 * memory areas, and then the tree is rebuilt again from the beginning.
69 * 2) KSM will only insert into the unstable tree, pages whose hash value
70 * has not changed since the previous scan of all memory areas.
71 * 3) The unstable tree is a RedBlack Tree - so its balancing is based on the
72 * colors of the nodes and not on their contents, assuring that even when
73 * the tree gets "corrupted" it won't get out of balance, so scanning time
74 * remains the same (also, searching and inserting nodes in an rbtree uses
75 * the same algorithm, so we have no overhead when we flush and rebuild).
76 * 4) KSM never flushes the stable tree, which means that even if it were to
77 * take 10 attempts to find a page in the unstable tree, once it is found,
78 * it is secured in the stable tree. (When we scan a new page, we first
79 * compare it against the stable tree, and then against the unstable tree.)
80 */
81
82/**
83 * struct mm_slot - ksm information per mm that is being scanned
84 * @link: link to the mm_slots hash list
85 * @mm_list: link into the mm_slots list, rooted in ksm_mm_head
Hugh Dickins6514d512009-12-14 17:59:19 -080086 * @rmap_list: head for this mm_slot's singly-linked list of rmap_items
Izik Eidus31dbd012009-09-21 17:02:03 -070087 * @mm: the mm that this information is valid for
88 */
89struct mm_slot {
90 struct hlist_node link;
91 struct list_head mm_list;
Hugh Dickins6514d512009-12-14 17:59:19 -080092 struct rmap_item *rmap_list;
Izik Eidus31dbd012009-09-21 17:02:03 -070093 struct mm_struct *mm;
94};
95
96/**
97 * struct ksm_scan - cursor for scanning
98 * @mm_slot: the current mm_slot we are scanning
99 * @address: the next address inside that to be scanned
Hugh Dickins6514d512009-12-14 17:59:19 -0800100 * @rmap_list: link to the next rmap to be scanned in the rmap_list
Izik Eidus31dbd012009-09-21 17:02:03 -0700101 * @seqnr: count of completed full scans (needed when removing unstable node)
102 *
103 * There is only the one ksm_scan instance of this cursor structure.
104 */
105struct ksm_scan {
106 struct mm_slot *mm_slot;
107 unsigned long address;
Hugh Dickins6514d512009-12-14 17:59:19 -0800108 struct rmap_item **rmap_list;
Izik Eidus31dbd012009-09-21 17:02:03 -0700109 unsigned long seqnr;
110};
111
112/**
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800113 * struct stable_node - node of the stable rbtree
114 * @node: rb node of this ksm page in the stable tree
115 * @hlist: hlist head of rmap_items using this ksm page
Hugh Dickins62b61f62009-12-14 17:59:33 -0800116 * @kpfn: page frame number of this ksm page
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800117 */
118struct stable_node {
119 struct rb_node node;
120 struct hlist_head hlist;
Hugh Dickins62b61f62009-12-14 17:59:33 -0800121 unsigned long kpfn;
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800122};
123
124/**
Izik Eidus31dbd012009-09-21 17:02:03 -0700125 * struct rmap_item - reverse mapping item for virtual addresses
Hugh Dickins6514d512009-12-14 17:59:19 -0800126 * @rmap_list: next rmap_item in mm_slot's singly-linked rmap_list
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800127 * @anon_vma: pointer to anon_vma for this mm,address, when in stable tree
Izik Eidus31dbd012009-09-21 17:02:03 -0700128 * @mm: the memory structure this rmap_item is pointing into
129 * @address: the virtual address this rmap_item tracks (+ flags in low bits)
130 * @oldchecksum: previous checksum of the page at that virtual address
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800131 * @node: rb node of this rmap_item in the unstable tree
132 * @head: pointer to stable_node heading this list in the stable tree
133 * @hlist: link into hlist of rmap_items hanging off that stable_node
Izik Eidus31dbd012009-09-21 17:02:03 -0700134 */
135struct rmap_item {
Hugh Dickins6514d512009-12-14 17:59:19 -0800136 struct rmap_item *rmap_list;
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800137 struct anon_vma *anon_vma; /* when stable */
Izik Eidus31dbd012009-09-21 17:02:03 -0700138 struct mm_struct *mm;
139 unsigned long address; /* + low bits used for flags below */
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800140 unsigned int oldchecksum; /* when unstable */
Izik Eidus31dbd012009-09-21 17:02:03 -0700141 union {
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800142 struct rb_node node; /* when node of unstable tree */
143 struct { /* when listed from stable tree */
144 struct stable_node *head;
145 struct hlist_node hlist;
146 };
Izik Eidus31dbd012009-09-21 17:02:03 -0700147 };
148};
149
150#define SEQNR_MASK 0x0ff /* low bits of unstable tree seqnr */
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800151#define UNSTABLE_FLAG 0x100 /* is a node of the unstable tree */
152#define STABLE_FLAG 0x200 /* is listed from the stable tree */
Izik Eidus31dbd012009-09-21 17:02:03 -0700153
154/* The stable and unstable tree heads */
155static struct rb_root root_stable_tree = RB_ROOT;
156static struct rb_root root_unstable_tree = RB_ROOT;
157
Lai Jiangshand9f89842010-08-09 17:20:02 -0700158#define MM_SLOTS_HASH_SHIFT 10
159#define MM_SLOTS_HASH_HEADS (1 << MM_SLOTS_HASH_SHIFT)
160static struct hlist_head mm_slots_hash[MM_SLOTS_HASH_HEADS];
Izik Eidus31dbd012009-09-21 17:02:03 -0700161
162static struct mm_slot ksm_mm_head = {
163 .mm_list = LIST_HEAD_INIT(ksm_mm_head.mm_list),
164};
165static struct ksm_scan ksm_scan = {
166 .mm_slot = &ksm_mm_head,
167};
168
169static struct kmem_cache *rmap_item_cache;
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800170static struct kmem_cache *stable_node_cache;
Izik Eidus31dbd012009-09-21 17:02:03 -0700171static struct kmem_cache *mm_slot_cache;
172
173/* The number of nodes in the stable tree */
Hugh Dickinsb4028262009-09-21 17:02:09 -0700174static unsigned long ksm_pages_shared;
Izik Eidus31dbd012009-09-21 17:02:03 -0700175
Hugh Dickinse178dfd2009-09-21 17:02:10 -0700176/* The number of page slots additionally sharing those nodes */
Hugh Dickinsb4028262009-09-21 17:02:09 -0700177static unsigned long ksm_pages_sharing;
Izik Eidus31dbd012009-09-21 17:02:03 -0700178
Hugh Dickins473b0ce2009-09-21 17:02:11 -0700179/* The number of nodes in the unstable tree */
180static unsigned long ksm_pages_unshared;
181
182/* The number of rmap_items in use: to calculate pages_volatile */
183static unsigned long ksm_rmap_items;
184
Izik Eidus31dbd012009-09-21 17:02:03 -0700185/* Number of pages ksmd should scan in one batch */
Izik Eidus2c6854f2009-09-23 15:56:04 -0700186static unsigned int ksm_thread_pages_to_scan = 100;
Izik Eidus31dbd012009-09-21 17:02:03 -0700187
188/* Milliseconds ksmd should sleep between batches */
Hugh Dickins2ffd8672009-09-21 17:02:23 -0700189static unsigned int ksm_thread_sleep_millisecs = 20;
Izik Eidus31dbd012009-09-21 17:02:03 -0700190
191#define KSM_RUN_STOP 0
192#define KSM_RUN_MERGE 1
193#define KSM_RUN_UNMERGE 2
Izik Eidus2c6854f2009-09-23 15:56:04 -0700194static unsigned int ksm_run = KSM_RUN_STOP;
Izik Eidus31dbd012009-09-21 17:02:03 -0700195
196static DECLARE_WAIT_QUEUE_HEAD(ksm_thread_wait);
197static DEFINE_MUTEX(ksm_thread_mutex);
198static DEFINE_SPINLOCK(ksm_mmlist_lock);
199
200#define KSM_KMEM_CACHE(__struct, __flags) kmem_cache_create("ksm_"#__struct,\
201 sizeof(struct __struct), __alignof__(struct __struct),\
202 (__flags), NULL)
203
204static int __init ksm_slab_init(void)
205{
206 rmap_item_cache = KSM_KMEM_CACHE(rmap_item, 0);
207 if (!rmap_item_cache)
208 goto out;
209
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800210 stable_node_cache = KSM_KMEM_CACHE(stable_node, 0);
211 if (!stable_node_cache)
212 goto out_free1;
213
Izik Eidus31dbd012009-09-21 17:02:03 -0700214 mm_slot_cache = KSM_KMEM_CACHE(mm_slot, 0);
215 if (!mm_slot_cache)
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800216 goto out_free2;
Izik Eidus31dbd012009-09-21 17:02:03 -0700217
218 return 0;
219
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800220out_free2:
221 kmem_cache_destroy(stable_node_cache);
222out_free1:
Izik Eidus31dbd012009-09-21 17:02:03 -0700223 kmem_cache_destroy(rmap_item_cache);
224out:
225 return -ENOMEM;
226}
227
228static void __init ksm_slab_free(void)
229{
230 kmem_cache_destroy(mm_slot_cache);
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800231 kmem_cache_destroy(stable_node_cache);
Izik Eidus31dbd012009-09-21 17:02:03 -0700232 kmem_cache_destroy(rmap_item_cache);
233 mm_slot_cache = NULL;
234}
235
236static inline struct rmap_item *alloc_rmap_item(void)
237{
Hugh Dickins473b0ce2009-09-21 17:02:11 -0700238 struct rmap_item *rmap_item;
239
240 rmap_item = kmem_cache_zalloc(rmap_item_cache, GFP_KERNEL);
241 if (rmap_item)
242 ksm_rmap_items++;
243 return rmap_item;
Izik Eidus31dbd012009-09-21 17:02:03 -0700244}
245
246static inline void free_rmap_item(struct rmap_item *rmap_item)
247{
Hugh Dickins473b0ce2009-09-21 17:02:11 -0700248 ksm_rmap_items--;
Izik Eidus31dbd012009-09-21 17:02:03 -0700249 rmap_item->mm = NULL; /* debug safety */
250 kmem_cache_free(rmap_item_cache, rmap_item);
251}
252
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800253static inline struct stable_node *alloc_stable_node(void)
254{
255 return kmem_cache_alloc(stable_node_cache, GFP_KERNEL);
256}
257
258static inline void free_stable_node(struct stable_node *stable_node)
259{
260 kmem_cache_free(stable_node_cache, stable_node);
261}
262
Izik Eidus31dbd012009-09-21 17:02:03 -0700263static inline struct mm_slot *alloc_mm_slot(void)
264{
265 if (!mm_slot_cache) /* initialization failed */
266 return NULL;
267 return kmem_cache_zalloc(mm_slot_cache, GFP_KERNEL);
268}
269
270static inline void free_mm_slot(struct mm_slot *mm_slot)
271{
272 kmem_cache_free(mm_slot_cache, mm_slot);
273}
274
Izik Eidus31dbd012009-09-21 17:02:03 -0700275static struct mm_slot *get_mm_slot(struct mm_struct *mm)
276{
277 struct mm_slot *mm_slot;
278 struct hlist_head *bucket;
279 struct hlist_node *node;
280
Lai Jiangshand9f89842010-08-09 17:20:02 -0700281 bucket = &mm_slots_hash[hash_ptr(mm, MM_SLOTS_HASH_SHIFT)];
Izik Eidus31dbd012009-09-21 17:02:03 -0700282 hlist_for_each_entry(mm_slot, node, bucket, link) {
283 if (mm == mm_slot->mm)
284 return mm_slot;
285 }
286 return NULL;
287}
288
289static void insert_to_mm_slots_hash(struct mm_struct *mm,
290 struct mm_slot *mm_slot)
291{
292 struct hlist_head *bucket;
293
Lai Jiangshand9f89842010-08-09 17:20:02 -0700294 bucket = &mm_slots_hash[hash_ptr(mm, MM_SLOTS_HASH_SHIFT)];
Izik Eidus31dbd012009-09-21 17:02:03 -0700295 mm_slot->mm = mm;
Izik Eidus31dbd012009-09-21 17:02:03 -0700296 hlist_add_head(&mm_slot->link, bucket);
297}
298
299static inline int in_stable_tree(struct rmap_item *rmap_item)
300{
301 return rmap_item->address & STABLE_FLAG;
302}
303
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800304static void hold_anon_vma(struct rmap_item *rmap_item,
305 struct anon_vma *anon_vma)
306{
307 rmap_item->anon_vma = anon_vma;
Rik van Riel76545062010-08-09 17:18:41 -0700308 get_anon_vma(anon_vma);
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800309}
310
Rik van Riel76545062010-08-09 17:18:41 -0700311static void ksm_drop_anon_vma(struct rmap_item *rmap_item)
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800312{
313 struct anon_vma *anon_vma = rmap_item->anon_vma;
314
Rik van Riel76545062010-08-09 17:18:41 -0700315 drop_anon_vma(anon_vma);
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800316}
317
Izik Eidus31dbd012009-09-21 17:02:03 -0700318/*
Hugh Dickinsa913e182009-09-21 17:02:26 -0700319 * ksmd, and unmerge_and_remove_all_rmap_items(), must not touch an mm's
320 * page tables after it has passed through ksm_exit() - which, if necessary,
321 * takes mmap_sem briefly to serialize against them. ksm_exit() does not set
322 * a special flag: they can just back out as soon as mm_users goes to zero.
323 * ksm_test_exit() is used throughout to make this test for exit: in some
324 * places for correctness, in some places just to avoid unnecessary work.
325 */
326static inline bool ksm_test_exit(struct mm_struct *mm)
327{
328 return atomic_read(&mm->mm_users) == 0;
329}
330
331/*
Izik Eidus31dbd012009-09-21 17:02:03 -0700332 * We use break_ksm to break COW on a ksm page: it's a stripped down
333 *
334 * if (get_user_pages(current, mm, addr, 1, 1, 1, &page, NULL) == 1)
335 * put_page(page);
336 *
337 * but taking great care only to touch a ksm page, in a VM_MERGEABLE vma,
338 * in case the application has unmapped and remapped mm,addr meanwhile.
339 * Could a ksm page appear anywhere else? Actually yes, in a VM_PFNMAP
340 * mmap of /dev/mem or /dev/kmem, where we would not want to touch it.
341 */
Hugh Dickinsd952b792009-09-21 17:02:16 -0700342static int break_ksm(struct vm_area_struct *vma, unsigned long addr)
Izik Eidus31dbd012009-09-21 17:02:03 -0700343{
344 struct page *page;
Hugh Dickinsd952b792009-09-21 17:02:16 -0700345 int ret = 0;
Izik Eidus31dbd012009-09-21 17:02:03 -0700346
347 do {
348 cond_resched();
349 page = follow_page(vma, addr, FOLL_GET);
Dan Carpenter22eccdd2010-04-23 13:18:10 -0400350 if (IS_ERR_OR_NULL(page))
Izik Eidus31dbd012009-09-21 17:02:03 -0700351 break;
352 if (PageKsm(page))
353 ret = handle_mm_fault(vma->vm_mm, vma, addr,
354 FAULT_FLAG_WRITE);
355 else
356 ret = VM_FAULT_WRITE;
357 put_page(page);
Hugh Dickinsd952b792009-09-21 17:02:16 -0700358 } while (!(ret & (VM_FAULT_WRITE | VM_FAULT_SIGBUS | VM_FAULT_OOM)));
359 /*
360 * We must loop because handle_mm_fault() may back out if there's
361 * any difficulty e.g. if pte accessed bit gets updated concurrently.
362 *
363 * VM_FAULT_WRITE is what we have been hoping for: it indicates that
364 * COW has been broken, even if the vma does not permit VM_WRITE;
365 * but note that a concurrent fault might break PageKsm for us.
366 *
367 * VM_FAULT_SIGBUS could occur if we race with truncation of the
368 * backing file, which also invalidates anonymous pages: that's
369 * okay, that truncation will have unmapped the PageKsm for us.
370 *
371 * VM_FAULT_OOM: at the time of writing (late July 2009), setting
372 * aside mem_cgroup limits, VM_FAULT_OOM would only be set if the
373 * current task has TIF_MEMDIE set, and will be OOM killed on return
374 * to user; and ksmd, having no mm, would never be chosen for that.
375 *
376 * But if the mm is in a limited mem_cgroup, then the fault may fail
377 * with VM_FAULT_OOM even if the current task is not TIF_MEMDIE; and
378 * even ksmd can fail in this way - though it's usually breaking ksm
379 * just to undo a merge it made a moment before, so unlikely to oom.
380 *
381 * That's a pity: we might therefore have more kernel pages allocated
382 * than we're counting as nodes in the stable tree; but ksm_do_scan
383 * will retry to break_cow on each pass, so should recover the page
384 * in due course. The important thing is to not let VM_MERGEABLE
385 * be cleared while any such pages might remain in the area.
386 */
387 return (ret & VM_FAULT_OOM) ? -ENOMEM : 0;
Izik Eidus31dbd012009-09-21 17:02:03 -0700388}
389
Hugh Dickins8dd35572009-12-14 17:59:18 -0800390static void break_cow(struct rmap_item *rmap_item)
Izik Eidus31dbd012009-09-21 17:02:03 -0700391{
Hugh Dickins8dd35572009-12-14 17:59:18 -0800392 struct mm_struct *mm = rmap_item->mm;
393 unsigned long addr = rmap_item->address;
Izik Eidus31dbd012009-09-21 17:02:03 -0700394 struct vm_area_struct *vma;
395
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800396 /*
397 * It is not an accident that whenever we want to break COW
398 * to undo, we also need to drop a reference to the anon_vma.
399 */
Rik van Riel76545062010-08-09 17:18:41 -0700400 ksm_drop_anon_vma(rmap_item);
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800401
Hugh Dickins81464e302009-09-21 17:02:15 -0700402 down_read(&mm->mmap_sem);
Hugh Dickins9ba69292009-09-21 17:02:20 -0700403 if (ksm_test_exit(mm))
404 goto out;
Izik Eidus31dbd012009-09-21 17:02:03 -0700405 vma = find_vma(mm, addr);
406 if (!vma || vma->vm_start > addr)
Hugh Dickins81464e302009-09-21 17:02:15 -0700407 goto out;
Izik Eidus31dbd012009-09-21 17:02:03 -0700408 if (!(vma->vm_flags & VM_MERGEABLE) || !vma->anon_vma)
Hugh Dickins81464e302009-09-21 17:02:15 -0700409 goto out;
Izik Eidus31dbd012009-09-21 17:02:03 -0700410 break_ksm(vma, addr);
Hugh Dickins81464e302009-09-21 17:02:15 -0700411out:
Izik Eidus31dbd012009-09-21 17:02:03 -0700412 up_read(&mm->mmap_sem);
413}
414
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800415static struct page *page_trans_compound_anon(struct page *page)
416{
417 if (PageTransCompound(page)) {
Andrea Arcangeli22e5c472011-01-13 15:47:20 -0800418 struct page *head = compound_trans_head(page);
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800419 /*
Andrea Arcangeli22e5c472011-01-13 15:47:20 -0800420 * head may actually be splitted and freed from under
421 * us but it's ok here.
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800422 */
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800423 if (PageAnon(head))
424 return head;
425 }
426 return NULL;
427}
428
Izik Eidus31dbd012009-09-21 17:02:03 -0700429static struct page *get_mergeable_page(struct rmap_item *rmap_item)
430{
431 struct mm_struct *mm = rmap_item->mm;
432 unsigned long addr = rmap_item->address;
433 struct vm_area_struct *vma;
434 struct page *page;
435
436 down_read(&mm->mmap_sem);
Hugh Dickins9ba69292009-09-21 17:02:20 -0700437 if (ksm_test_exit(mm))
438 goto out;
Izik Eidus31dbd012009-09-21 17:02:03 -0700439 vma = find_vma(mm, addr);
440 if (!vma || vma->vm_start > addr)
441 goto out;
442 if (!(vma->vm_flags & VM_MERGEABLE) || !vma->anon_vma)
443 goto out;
444
445 page = follow_page(vma, addr, FOLL_GET);
Dan Carpenter22eccdd2010-04-23 13:18:10 -0400446 if (IS_ERR_OR_NULL(page))
Izik Eidus31dbd012009-09-21 17:02:03 -0700447 goto out;
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800448 if (PageAnon(page) || page_trans_compound_anon(page)) {
Izik Eidus31dbd012009-09-21 17:02:03 -0700449 flush_anon_page(vma, page, addr);
450 flush_dcache_page(page);
451 } else {
452 put_page(page);
453out: page = NULL;
454 }
455 up_read(&mm->mmap_sem);
456 return page;
457}
458
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800459static void remove_node_from_stable_tree(struct stable_node *stable_node)
460{
461 struct rmap_item *rmap_item;
462 struct hlist_node *hlist;
463
464 hlist_for_each_entry(rmap_item, hlist, &stable_node->hlist, hlist) {
465 if (rmap_item->hlist.next)
466 ksm_pages_sharing--;
467 else
468 ksm_pages_shared--;
Rik van Riel76545062010-08-09 17:18:41 -0700469 ksm_drop_anon_vma(rmap_item);
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800470 rmap_item->address &= PAGE_MASK;
471 cond_resched();
472 }
473
474 rb_erase(&stable_node->node, &root_stable_tree);
475 free_stable_node(stable_node);
476}
477
478/*
479 * get_ksm_page: checks if the page indicated by the stable node
480 * is still its ksm page, despite having held no reference to it.
481 * In which case we can trust the content of the page, and it
482 * returns the gotten page; but if the page has now been zapped,
483 * remove the stale node from the stable tree and return NULL.
484 *
485 * You would expect the stable_node to hold a reference to the ksm page.
486 * But if it increments the page's count, swapping out has to wait for
487 * ksmd to come around again before it can free the page, which may take
488 * seconds or even minutes: much too unresponsive. So instead we use a
489 * "keyhole reference": access to the ksm page from the stable node peeps
490 * out through its keyhole to see if that page still holds the right key,
491 * pointing back to this stable node. This relies on freeing a PageAnon
492 * page to reset its page->mapping to NULL, and relies on no other use of
493 * a page to put something that might look like our key in page->mapping.
494 *
495 * include/linux/pagemap.h page_cache_get_speculative() is a good reference,
496 * but this is different - made simpler by ksm_thread_mutex being held, but
497 * interesting for assuming that no other use of the struct page could ever
498 * put our expected_mapping into page->mapping (or a field of the union which
499 * coincides with page->mapping). The RCU calls are not for KSM at all, but
500 * to keep the page_count protocol described with page_cache_get_speculative.
501 *
502 * Note: it is possible that get_ksm_page() will return NULL one moment,
503 * then page the next, if the page is in between page_freeze_refs() and
504 * page_unfreeze_refs(): this shouldn't be a problem anywhere, the page
505 * is on its way to being freed; but it is an anomaly to bear in mind.
506 */
507static struct page *get_ksm_page(struct stable_node *stable_node)
508{
509 struct page *page;
510 void *expected_mapping;
511
Hugh Dickins62b61f62009-12-14 17:59:33 -0800512 page = pfn_to_page(stable_node->kpfn);
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800513 expected_mapping = (void *)stable_node +
514 (PAGE_MAPPING_ANON | PAGE_MAPPING_KSM);
515 rcu_read_lock();
516 if (page->mapping != expected_mapping)
517 goto stale;
518 if (!get_page_unless_zero(page))
519 goto stale;
520 if (page->mapping != expected_mapping) {
521 put_page(page);
522 goto stale;
523 }
524 rcu_read_unlock();
525 return page;
526stale:
527 rcu_read_unlock();
528 remove_node_from_stable_tree(stable_node);
529 return NULL;
530}
531
Izik Eidus31dbd012009-09-21 17:02:03 -0700532/*
Izik Eidus31dbd012009-09-21 17:02:03 -0700533 * Removing rmap_item from stable or unstable tree.
534 * This function will clean the information from the stable/unstable tree.
535 */
536static void remove_rmap_item_from_tree(struct rmap_item *rmap_item)
537{
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800538 if (rmap_item->address & STABLE_FLAG) {
539 struct stable_node *stable_node;
Hugh Dickins5ad64682009-12-14 17:59:24 -0800540 struct page *page;
Izik Eidus31dbd012009-09-21 17:02:03 -0700541
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800542 stable_node = rmap_item->head;
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800543 page = get_ksm_page(stable_node);
544 if (!page)
545 goto out;
546
Hugh Dickins5ad64682009-12-14 17:59:24 -0800547 lock_page(page);
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800548 hlist_del(&rmap_item->hlist);
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800549 unlock_page(page);
550 put_page(page);
Hugh Dickins08beca42009-12-14 17:59:21 -0800551
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800552 if (stable_node->hlist.first)
553 ksm_pages_sharing--;
554 else
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800555 ksm_pages_shared--;
Izik Eidus31dbd012009-09-21 17:02:03 -0700556
Rik van Riel76545062010-08-09 17:18:41 -0700557 ksm_drop_anon_vma(rmap_item);
Hugh Dickins93d17712009-12-14 17:59:16 -0800558 rmap_item->address &= PAGE_MASK;
Izik Eidus31dbd012009-09-21 17:02:03 -0700559
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800560 } else if (rmap_item->address & UNSTABLE_FLAG) {
Izik Eidus31dbd012009-09-21 17:02:03 -0700561 unsigned char age;
562 /*
Hugh Dickins9ba69292009-09-21 17:02:20 -0700563 * Usually ksmd can and must skip the rb_erase, because
Izik Eidus31dbd012009-09-21 17:02:03 -0700564 * root_unstable_tree was already reset to RB_ROOT.
Hugh Dickins9ba69292009-09-21 17:02:20 -0700565 * But be careful when an mm is exiting: do the rb_erase
566 * if this rmap_item was inserted by this scan, rather
567 * than left over from before.
Izik Eidus31dbd012009-09-21 17:02:03 -0700568 */
569 age = (unsigned char)(ksm_scan.seqnr - rmap_item->address);
Hugh Dickinscd551f92009-09-21 17:02:17 -0700570 BUG_ON(age > 1);
Izik Eidus31dbd012009-09-21 17:02:03 -0700571 if (!age)
572 rb_erase(&rmap_item->node, &root_unstable_tree);
Izik Eidus31dbd012009-09-21 17:02:03 -0700573
Hugh Dickins93d17712009-12-14 17:59:16 -0800574 ksm_pages_unshared--;
575 rmap_item->address &= PAGE_MASK;
576 }
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800577out:
Izik Eidus31dbd012009-09-21 17:02:03 -0700578 cond_resched(); /* we're called from many long loops */
579}
580
Izik Eidus31dbd012009-09-21 17:02:03 -0700581static void remove_trailing_rmap_items(struct mm_slot *mm_slot,
Hugh Dickins6514d512009-12-14 17:59:19 -0800582 struct rmap_item **rmap_list)
Izik Eidus31dbd012009-09-21 17:02:03 -0700583{
Hugh Dickins6514d512009-12-14 17:59:19 -0800584 while (*rmap_list) {
585 struct rmap_item *rmap_item = *rmap_list;
586 *rmap_list = rmap_item->rmap_list;
Izik Eidus31dbd012009-09-21 17:02:03 -0700587 remove_rmap_item_from_tree(rmap_item);
Izik Eidus31dbd012009-09-21 17:02:03 -0700588 free_rmap_item(rmap_item);
589 }
590}
591
592/*
593 * Though it's very tempting to unmerge in_stable_tree(rmap_item)s rather
594 * than check every pte of a given vma, the locking doesn't quite work for
595 * that - an rmap_item is assigned to the stable tree after inserting ksm
596 * page and upping mmap_sem. Nor does it fit with the way we skip dup'ing
597 * rmap_items from parent to child at fork time (so as not to waste time
598 * if exit comes before the next scan reaches it).
Hugh Dickins81464e302009-09-21 17:02:15 -0700599 *
600 * Similarly, although we'd like to remove rmap_items (so updating counts
601 * and freeing memory) when unmerging an area, it's easier to leave that
602 * to the next pass of ksmd - consider, for example, how ksmd might be
603 * in cmp_and_merge_page on one of the rmap_items we would be removing.
Izik Eidus31dbd012009-09-21 17:02:03 -0700604 */
Hugh Dickinsd952b792009-09-21 17:02:16 -0700605static int unmerge_ksm_pages(struct vm_area_struct *vma,
606 unsigned long start, unsigned long end)
Izik Eidus31dbd012009-09-21 17:02:03 -0700607{
608 unsigned long addr;
Hugh Dickinsd952b792009-09-21 17:02:16 -0700609 int err = 0;
Izik Eidus31dbd012009-09-21 17:02:03 -0700610
Hugh Dickinsd952b792009-09-21 17:02:16 -0700611 for (addr = start; addr < end && !err; addr += PAGE_SIZE) {
Hugh Dickins9ba69292009-09-21 17:02:20 -0700612 if (ksm_test_exit(vma->vm_mm))
613 break;
Hugh Dickinsd952b792009-09-21 17:02:16 -0700614 if (signal_pending(current))
615 err = -ERESTARTSYS;
616 else
617 err = break_ksm(vma, addr);
618 }
619 return err;
Izik Eidus31dbd012009-09-21 17:02:03 -0700620}
621
Hugh Dickins2ffd8672009-09-21 17:02:23 -0700622#ifdef CONFIG_SYSFS
623/*
624 * Only called through the sysfs control interface:
625 */
Hugh Dickinsd952b792009-09-21 17:02:16 -0700626static int unmerge_and_remove_all_rmap_items(void)
Izik Eidus31dbd012009-09-21 17:02:03 -0700627{
628 struct mm_slot *mm_slot;
629 struct mm_struct *mm;
630 struct vm_area_struct *vma;
Hugh Dickinsd952b792009-09-21 17:02:16 -0700631 int err = 0;
Izik Eidus31dbd012009-09-21 17:02:03 -0700632
Hugh Dickinsd952b792009-09-21 17:02:16 -0700633 spin_lock(&ksm_mmlist_lock);
Hugh Dickins9ba69292009-09-21 17:02:20 -0700634 ksm_scan.mm_slot = list_entry(ksm_mm_head.mm_list.next,
Hugh Dickinsd952b792009-09-21 17:02:16 -0700635 struct mm_slot, mm_list);
636 spin_unlock(&ksm_mmlist_lock);
637
Hugh Dickins9ba69292009-09-21 17:02:20 -0700638 for (mm_slot = ksm_scan.mm_slot;
639 mm_slot != &ksm_mm_head; mm_slot = ksm_scan.mm_slot) {
Izik Eidus31dbd012009-09-21 17:02:03 -0700640 mm = mm_slot->mm;
641 down_read(&mm->mmap_sem);
642 for (vma = mm->mmap; vma; vma = vma->vm_next) {
Hugh Dickins9ba69292009-09-21 17:02:20 -0700643 if (ksm_test_exit(mm))
644 break;
Izik Eidus31dbd012009-09-21 17:02:03 -0700645 if (!(vma->vm_flags & VM_MERGEABLE) || !vma->anon_vma)
646 continue;
Hugh Dickinsd952b792009-09-21 17:02:16 -0700647 err = unmerge_ksm_pages(vma,
648 vma->vm_start, vma->vm_end);
Hugh Dickins9ba69292009-09-21 17:02:20 -0700649 if (err)
650 goto error;
Izik Eidus31dbd012009-09-21 17:02:03 -0700651 }
Hugh Dickins9ba69292009-09-21 17:02:20 -0700652
Hugh Dickins6514d512009-12-14 17:59:19 -0800653 remove_trailing_rmap_items(mm_slot, &mm_slot->rmap_list);
Hugh Dickinsd952b792009-09-21 17:02:16 -0700654
655 spin_lock(&ksm_mmlist_lock);
Hugh Dickins9ba69292009-09-21 17:02:20 -0700656 ksm_scan.mm_slot = list_entry(mm_slot->mm_list.next,
Hugh Dickinsd952b792009-09-21 17:02:16 -0700657 struct mm_slot, mm_list);
Hugh Dickins9ba69292009-09-21 17:02:20 -0700658 if (ksm_test_exit(mm)) {
659 hlist_del(&mm_slot->link);
660 list_del(&mm_slot->mm_list);
661 spin_unlock(&ksm_mmlist_lock);
662
663 free_mm_slot(mm_slot);
664 clear_bit(MMF_VM_MERGEABLE, &mm->flags);
665 up_read(&mm->mmap_sem);
666 mmdrop(mm);
667 } else {
668 spin_unlock(&ksm_mmlist_lock);
669 up_read(&mm->mmap_sem);
670 }
Izik Eidus31dbd012009-09-21 17:02:03 -0700671 }
672
Hugh Dickinsd952b792009-09-21 17:02:16 -0700673 ksm_scan.seqnr = 0;
Hugh Dickins9ba69292009-09-21 17:02:20 -0700674 return 0;
675
676error:
677 up_read(&mm->mmap_sem);
Izik Eidus31dbd012009-09-21 17:02:03 -0700678 spin_lock(&ksm_mmlist_lock);
Hugh Dickinsd952b792009-09-21 17:02:16 -0700679 ksm_scan.mm_slot = &ksm_mm_head;
Izik Eidus31dbd012009-09-21 17:02:03 -0700680 spin_unlock(&ksm_mmlist_lock);
Hugh Dickinsd952b792009-09-21 17:02:16 -0700681 return err;
Izik Eidus31dbd012009-09-21 17:02:03 -0700682}
Hugh Dickins2ffd8672009-09-21 17:02:23 -0700683#endif /* CONFIG_SYSFS */
Izik Eidus31dbd012009-09-21 17:02:03 -0700684
Izik Eidus31dbd012009-09-21 17:02:03 -0700685static u32 calc_checksum(struct page *page)
686{
687 u32 checksum;
688 void *addr = kmap_atomic(page, KM_USER0);
689 checksum = jhash2(addr, PAGE_SIZE / 4, 17);
690 kunmap_atomic(addr, KM_USER0);
691 return checksum;
692}
693
694static int memcmp_pages(struct page *page1, struct page *page2)
695{
696 char *addr1, *addr2;
697 int ret;
698
699 addr1 = kmap_atomic(page1, KM_USER0);
700 addr2 = kmap_atomic(page2, KM_USER1);
701 ret = memcmp(addr1, addr2, PAGE_SIZE);
702 kunmap_atomic(addr2, KM_USER1);
703 kunmap_atomic(addr1, KM_USER0);
704 return ret;
705}
706
707static inline int pages_identical(struct page *page1, struct page *page2)
708{
709 return !memcmp_pages(page1, page2);
710}
711
712static int write_protect_page(struct vm_area_struct *vma, struct page *page,
713 pte_t *orig_pte)
714{
715 struct mm_struct *mm = vma->vm_mm;
716 unsigned long addr;
717 pte_t *ptep;
718 spinlock_t *ptl;
719 int swapped;
720 int err = -EFAULT;
721
722 addr = page_address_in_vma(page, vma);
723 if (addr == -EFAULT)
724 goto out;
725
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800726 BUG_ON(PageTransCompound(page));
Izik Eidus31dbd012009-09-21 17:02:03 -0700727 ptep = page_check_address(page, mm, addr, &ptl, 0);
728 if (!ptep)
729 goto out;
730
Hugh Dickins4e316352010-10-02 17:49:08 -0700731 if (pte_write(*ptep) || pte_dirty(*ptep)) {
Izik Eidus31dbd012009-09-21 17:02:03 -0700732 pte_t entry;
733
734 swapped = PageSwapCache(page);
735 flush_cache_page(vma, addr, page_to_pfn(page));
736 /*
737 * Ok this is tricky, when get_user_pages_fast() run it doesnt
738 * take any lock, therefore the check that we are going to make
739 * with the pagecount against the mapcount is racey and
740 * O_DIRECT can happen right after the check.
741 * So we clear the pte and flush the tlb before the check
742 * this assure us that no O_DIRECT can happen after the check
743 * or in the middle of the check.
744 */
745 entry = ptep_clear_flush(vma, addr, ptep);
746 /*
747 * Check that no O_DIRECT or similar I/O is in progress on the
748 * page
749 */
Hugh Dickins31e855e2009-12-14 17:59:17 -0800750 if (page_mapcount(page) + 1 + swapped != page_count(page)) {
Robin Holtcb532372010-03-23 13:35:26 -0700751 set_pte_at(mm, addr, ptep, entry);
Izik Eidus31dbd012009-09-21 17:02:03 -0700752 goto out_unlock;
753 }
Hugh Dickins4e316352010-10-02 17:49:08 -0700754 if (pte_dirty(entry))
755 set_page_dirty(page);
756 entry = pte_mkclean(pte_wrprotect(entry));
Izik Eidus31dbd012009-09-21 17:02:03 -0700757 set_pte_at_notify(mm, addr, ptep, entry);
758 }
759 *orig_pte = *ptep;
760 err = 0;
761
762out_unlock:
763 pte_unmap_unlock(ptep, ptl);
764out:
765 return err;
766}
767
768/**
769 * replace_page - replace page in vma by new ksm page
Hugh Dickins8dd35572009-12-14 17:59:18 -0800770 * @vma: vma that holds the pte pointing to page
771 * @page: the page we are replacing by kpage
772 * @kpage: the ksm page we replace page by
Izik Eidus31dbd012009-09-21 17:02:03 -0700773 * @orig_pte: the original value of the pte
774 *
775 * Returns 0 on success, -EFAULT on failure.
776 */
Hugh Dickins8dd35572009-12-14 17:59:18 -0800777static int replace_page(struct vm_area_struct *vma, struct page *page,
778 struct page *kpage, pte_t orig_pte)
Izik Eidus31dbd012009-09-21 17:02:03 -0700779{
780 struct mm_struct *mm = vma->vm_mm;
781 pgd_t *pgd;
782 pud_t *pud;
783 pmd_t *pmd;
784 pte_t *ptep;
785 spinlock_t *ptl;
786 unsigned long addr;
Izik Eidus31dbd012009-09-21 17:02:03 -0700787 int err = -EFAULT;
788
Hugh Dickins8dd35572009-12-14 17:59:18 -0800789 addr = page_address_in_vma(page, vma);
Izik Eidus31dbd012009-09-21 17:02:03 -0700790 if (addr == -EFAULT)
791 goto out;
792
793 pgd = pgd_offset(mm, addr);
794 if (!pgd_present(*pgd))
795 goto out;
796
797 pud = pud_offset(pgd, addr);
798 if (!pud_present(*pud))
799 goto out;
800
801 pmd = pmd_offset(pud, addr);
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800802 BUG_ON(pmd_trans_huge(*pmd));
Izik Eidus31dbd012009-09-21 17:02:03 -0700803 if (!pmd_present(*pmd))
804 goto out;
805
806 ptep = pte_offset_map_lock(mm, pmd, addr, &ptl);
807 if (!pte_same(*ptep, orig_pte)) {
808 pte_unmap_unlock(ptep, ptl);
809 goto out;
810 }
811
Hugh Dickins8dd35572009-12-14 17:59:18 -0800812 get_page(kpage);
Hugh Dickins5ad64682009-12-14 17:59:24 -0800813 page_add_anon_rmap(kpage, vma, addr);
Izik Eidus31dbd012009-09-21 17:02:03 -0700814
815 flush_cache_page(vma, addr, pte_pfn(*ptep));
816 ptep_clear_flush(vma, addr, ptep);
Hugh Dickins8dd35572009-12-14 17:59:18 -0800817 set_pte_at_notify(mm, addr, ptep, mk_pte(kpage, vma->vm_page_prot));
Izik Eidus31dbd012009-09-21 17:02:03 -0700818
Hugh Dickins8dd35572009-12-14 17:59:18 -0800819 page_remove_rmap(page);
Hugh Dickinsae52a2a2011-01-13 15:46:28 -0800820 if (!page_mapped(page))
821 try_to_free_swap(page);
Hugh Dickins8dd35572009-12-14 17:59:18 -0800822 put_page(page);
Izik Eidus31dbd012009-09-21 17:02:03 -0700823
824 pte_unmap_unlock(ptep, ptl);
825 err = 0;
826out:
827 return err;
828}
829
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800830static int page_trans_compound_anon_split(struct page *page)
831{
832 int ret = 0;
833 struct page *transhuge_head = page_trans_compound_anon(page);
834 if (transhuge_head) {
835 /* Get the reference on the head to split it. */
836 if (get_page_unless_zero(transhuge_head)) {
837 /*
838 * Recheck we got the reference while the head
839 * was still anonymous.
840 */
841 if (PageAnon(transhuge_head))
842 ret = split_huge_page(transhuge_head);
843 else
844 /*
845 * Retry later if split_huge_page run
846 * from under us.
847 */
848 ret = 1;
849 put_page(transhuge_head);
850 } else
851 /* Retry later if split_huge_page run from under us. */
852 ret = 1;
853 }
854 return ret;
855}
856
Izik Eidus31dbd012009-09-21 17:02:03 -0700857/*
858 * try_to_merge_one_page - take two pages and merge them into one
Hugh Dickins8dd35572009-12-14 17:59:18 -0800859 * @vma: the vma that holds the pte pointing to page
860 * @page: the PageAnon page that we want to replace with kpage
Hugh Dickins80e148222009-12-14 17:59:29 -0800861 * @kpage: the PageKsm page that we want to map instead of page,
862 * or NULL the first time when we want to use page as kpage.
Izik Eidus31dbd012009-09-21 17:02:03 -0700863 *
864 * This function returns 0 if the pages were merged, -EFAULT otherwise.
865 */
866static int try_to_merge_one_page(struct vm_area_struct *vma,
Hugh Dickins8dd35572009-12-14 17:59:18 -0800867 struct page *page, struct page *kpage)
Izik Eidus31dbd012009-09-21 17:02:03 -0700868{
869 pte_t orig_pte = __pte(0);
870 int err = -EFAULT;
871
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800872 if (page == kpage) /* ksm page forked */
873 return 0;
874
Izik Eidus31dbd012009-09-21 17:02:03 -0700875 if (!(vma->vm_flags & VM_MERGEABLE))
876 goto out;
Andrea Arcangeli29ad7682011-01-13 15:47:19 -0800877 if (PageTransCompound(page) && page_trans_compound_anon_split(page))
878 goto out;
879 BUG_ON(PageTransCompound(page));
Hugh Dickins8dd35572009-12-14 17:59:18 -0800880 if (!PageAnon(page))
Izik Eidus31dbd012009-09-21 17:02:03 -0700881 goto out;
882
Izik Eidus31dbd012009-09-21 17:02:03 -0700883 /*
884 * We need the page lock to read a stable PageSwapCache in
885 * write_protect_page(). We use trylock_page() instead of
886 * lock_page() because we don't want to wait here - we
887 * prefer to continue scanning and merging different pages,
888 * then come back to this page when it is unlocked.
889 */
Hugh Dickins8dd35572009-12-14 17:59:18 -0800890 if (!trylock_page(page))
Hugh Dickins31e855e2009-12-14 17:59:17 -0800891 goto out;
Izik Eidus31dbd012009-09-21 17:02:03 -0700892 /*
893 * If this anonymous page is mapped only here, its pte may need
894 * to be write-protected. If it's mapped elsewhere, all of its
895 * ptes are necessarily already write-protected. But in either
896 * case, we need to lock and check page_count is not raised.
897 */
Hugh Dickins80e148222009-12-14 17:59:29 -0800898 if (write_protect_page(vma, page, &orig_pte) == 0) {
899 if (!kpage) {
900 /*
901 * While we hold page lock, upgrade page from
902 * PageAnon+anon_vma to PageKsm+NULL stable_node:
903 * stable_tree_insert() will update stable_node.
904 */
905 set_page_stable_node(page, NULL);
906 mark_page_accessed(page);
907 err = 0;
908 } else if (pages_identical(page, kpage))
909 err = replace_page(vma, page, kpage, orig_pte);
910 }
Izik Eidus31dbd012009-09-21 17:02:03 -0700911
Hugh Dickins80e148222009-12-14 17:59:29 -0800912 if ((vma->vm_flags & VM_LOCKED) && kpage && !err) {
Hugh Dickins73848b42009-12-14 17:59:22 -0800913 munlock_vma_page(page);
Hugh Dickins5ad64682009-12-14 17:59:24 -0800914 if (!PageMlocked(kpage)) {
915 unlock_page(page);
Hugh Dickins5ad64682009-12-14 17:59:24 -0800916 lock_page(kpage);
917 mlock_vma_page(kpage);
918 page = kpage; /* for final unlock */
919 }
920 }
Hugh Dickins73848b42009-12-14 17:59:22 -0800921
Hugh Dickins8dd35572009-12-14 17:59:18 -0800922 unlock_page(page);
Izik Eidus31dbd012009-09-21 17:02:03 -0700923out:
924 return err;
925}
926
927/*
Hugh Dickins81464e302009-09-21 17:02:15 -0700928 * try_to_merge_with_ksm_page - like try_to_merge_two_pages,
929 * but no new kernel page is allocated: kpage must already be a ksm page.
Hugh Dickins8dd35572009-12-14 17:59:18 -0800930 *
931 * This function returns 0 if the pages were merged, -EFAULT otherwise.
Hugh Dickins81464e302009-09-21 17:02:15 -0700932 */
Hugh Dickins8dd35572009-12-14 17:59:18 -0800933static int try_to_merge_with_ksm_page(struct rmap_item *rmap_item,
934 struct page *page, struct page *kpage)
Hugh Dickins81464e302009-09-21 17:02:15 -0700935{
Hugh Dickins8dd35572009-12-14 17:59:18 -0800936 struct mm_struct *mm = rmap_item->mm;
Hugh Dickins81464e302009-09-21 17:02:15 -0700937 struct vm_area_struct *vma;
938 int err = -EFAULT;
939
Hugh Dickins8dd35572009-12-14 17:59:18 -0800940 down_read(&mm->mmap_sem);
941 if (ksm_test_exit(mm))
942 goto out;
943 vma = find_vma(mm, rmap_item->address);
944 if (!vma || vma->vm_start > rmap_item->address)
Hugh Dickins9ba69292009-09-21 17:02:20 -0700945 goto out;
946
Hugh Dickins8dd35572009-12-14 17:59:18 -0800947 err = try_to_merge_one_page(vma, page, kpage);
Hugh Dickinsdb114b82009-12-14 17:59:25 -0800948 if (err)
949 goto out;
950
951 /* Must get reference to anon_vma while still holding mmap_sem */
952 hold_anon_vma(rmap_item, vma->anon_vma);
Hugh Dickins81464e302009-09-21 17:02:15 -0700953out:
Hugh Dickins8dd35572009-12-14 17:59:18 -0800954 up_read(&mm->mmap_sem);
Hugh Dickins81464e302009-09-21 17:02:15 -0700955 return err;
956}
957
958/*
Izik Eidus31dbd012009-09-21 17:02:03 -0700959 * try_to_merge_two_pages - take two identical pages and prepare them
960 * to be merged into one page.
961 *
Hugh Dickins8dd35572009-12-14 17:59:18 -0800962 * This function returns the kpage if we successfully merged two identical
963 * pages into one ksm page, NULL otherwise.
Izik Eidus31dbd012009-09-21 17:02:03 -0700964 *
Hugh Dickins80e148222009-12-14 17:59:29 -0800965 * Note that this function upgrades page to ksm page: if one of the pages
Izik Eidus31dbd012009-09-21 17:02:03 -0700966 * is already a ksm page, try_to_merge_with_ksm_page should be used.
967 */
Hugh Dickins8dd35572009-12-14 17:59:18 -0800968static struct page *try_to_merge_two_pages(struct rmap_item *rmap_item,
969 struct page *page,
970 struct rmap_item *tree_rmap_item,
971 struct page *tree_page)
Izik Eidus31dbd012009-09-21 17:02:03 -0700972{
Hugh Dickins80e148222009-12-14 17:59:29 -0800973 int err;
Izik Eidus31dbd012009-09-21 17:02:03 -0700974
Hugh Dickins80e148222009-12-14 17:59:29 -0800975 err = try_to_merge_with_ksm_page(rmap_item, page, NULL);
Izik Eidus31dbd012009-09-21 17:02:03 -0700976 if (!err) {
Hugh Dickins8dd35572009-12-14 17:59:18 -0800977 err = try_to_merge_with_ksm_page(tree_rmap_item,
Hugh Dickins80e148222009-12-14 17:59:29 -0800978 tree_page, page);
Izik Eidus31dbd012009-09-21 17:02:03 -0700979 /*
Hugh Dickins81464e302009-09-21 17:02:15 -0700980 * If that fails, we have a ksm page with only one pte
981 * pointing to it: so break it.
Izik Eidus31dbd012009-09-21 17:02:03 -0700982 */
Hugh Dickins4035c07a2009-12-14 17:59:27 -0800983 if (err)
Hugh Dickins8dd35572009-12-14 17:59:18 -0800984 break_cow(rmap_item);
Izik Eidus31dbd012009-09-21 17:02:03 -0700985 }
Hugh Dickins80e148222009-12-14 17:59:29 -0800986 return err ? NULL : page;
Izik Eidus31dbd012009-09-21 17:02:03 -0700987}
988
989/*
Hugh Dickins8dd35572009-12-14 17:59:18 -0800990 * stable_tree_search - search for page inside the stable tree
Izik Eidus31dbd012009-09-21 17:02:03 -0700991 *
992 * This function checks if there is a page inside the stable tree
993 * with identical content to the page that we are scanning right now.
994 *
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -0800995 * This function returns the stable tree node of identical content if found,
Izik Eidus31dbd012009-09-21 17:02:03 -0700996 * NULL otherwise.
997 */
Hugh Dickins62b61f62009-12-14 17:59:33 -0800998static struct page *stable_tree_search(struct page *page)
Izik Eidus31dbd012009-09-21 17:02:03 -0700999{
1000 struct rb_node *node = root_stable_tree.rb_node;
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001001 struct stable_node *stable_node;
Izik Eidus31dbd012009-09-21 17:02:03 -07001002
Hugh Dickins08beca42009-12-14 17:59:21 -08001003 stable_node = page_stable_node(page);
1004 if (stable_node) { /* ksm page forked */
1005 get_page(page);
Hugh Dickins62b61f62009-12-14 17:59:33 -08001006 return page;
Hugh Dickins08beca42009-12-14 17:59:21 -08001007 }
1008
Izik Eidus31dbd012009-09-21 17:02:03 -07001009 while (node) {
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001010 struct page *tree_page;
Izik Eidus31dbd012009-09-21 17:02:03 -07001011 int ret;
1012
Hugh Dickins08beca42009-12-14 17:59:21 -08001013 cond_resched();
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001014 stable_node = rb_entry(node, struct stable_node, node);
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001015 tree_page = get_ksm_page(stable_node);
1016 if (!tree_page)
1017 return NULL;
Izik Eidus31dbd012009-09-21 17:02:03 -07001018
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001019 ret = memcmp_pages(page, tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001020
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001021 if (ret < 0) {
1022 put_page(tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001023 node = node->rb_left;
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001024 } else if (ret > 0) {
1025 put_page(tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001026 node = node->rb_right;
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001027 } else
Hugh Dickins62b61f62009-12-14 17:59:33 -08001028 return tree_page;
Izik Eidus31dbd012009-09-21 17:02:03 -07001029 }
1030
1031 return NULL;
1032}
1033
1034/*
1035 * stable_tree_insert - insert rmap_item pointing to new ksm page
1036 * into the stable tree.
1037 *
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001038 * This function returns the stable tree node just allocated on success,
1039 * NULL otherwise.
Izik Eidus31dbd012009-09-21 17:02:03 -07001040 */
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001041static struct stable_node *stable_tree_insert(struct page *kpage)
Izik Eidus31dbd012009-09-21 17:02:03 -07001042{
1043 struct rb_node **new = &root_stable_tree.rb_node;
1044 struct rb_node *parent = NULL;
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001045 struct stable_node *stable_node;
Izik Eidus31dbd012009-09-21 17:02:03 -07001046
1047 while (*new) {
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001048 struct page *tree_page;
Izik Eidus31dbd012009-09-21 17:02:03 -07001049 int ret;
1050
Hugh Dickins08beca42009-12-14 17:59:21 -08001051 cond_resched();
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001052 stable_node = rb_entry(*new, struct stable_node, node);
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001053 tree_page = get_ksm_page(stable_node);
1054 if (!tree_page)
1055 return NULL;
Izik Eidus31dbd012009-09-21 17:02:03 -07001056
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001057 ret = memcmp_pages(kpage, tree_page);
1058 put_page(tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001059
1060 parent = *new;
1061 if (ret < 0)
1062 new = &parent->rb_left;
1063 else if (ret > 0)
1064 new = &parent->rb_right;
1065 else {
1066 /*
1067 * It is not a bug that stable_tree_search() didn't
1068 * find this node: because at that time our page was
1069 * not yet write-protected, so may have changed since.
1070 */
1071 return NULL;
1072 }
1073 }
1074
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001075 stable_node = alloc_stable_node();
1076 if (!stable_node)
1077 return NULL;
Izik Eidus31dbd012009-09-21 17:02:03 -07001078
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001079 rb_link_node(&stable_node->node, parent, new);
1080 rb_insert_color(&stable_node->node, &root_stable_tree);
1081
1082 INIT_HLIST_HEAD(&stable_node->hlist);
1083
Hugh Dickins62b61f62009-12-14 17:59:33 -08001084 stable_node->kpfn = page_to_pfn(kpage);
Hugh Dickins08beca42009-12-14 17:59:21 -08001085 set_page_stable_node(kpage, stable_node);
1086
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001087 return stable_node;
Izik Eidus31dbd012009-09-21 17:02:03 -07001088}
1089
1090/*
Hugh Dickins8dd35572009-12-14 17:59:18 -08001091 * unstable_tree_search_insert - search for identical page,
1092 * else insert rmap_item into the unstable tree.
Izik Eidus31dbd012009-09-21 17:02:03 -07001093 *
1094 * This function searches for a page in the unstable tree identical to the
1095 * page currently being scanned; and if no identical page is found in the
1096 * tree, we insert rmap_item as a new object into the unstable tree.
1097 *
1098 * This function returns pointer to rmap_item found to be identical
1099 * to the currently scanned page, NULL otherwise.
1100 *
1101 * This function does both searching and inserting, because they share
1102 * the same walking algorithm in an rbtree.
1103 */
Hugh Dickins8dd35572009-12-14 17:59:18 -08001104static
1105struct rmap_item *unstable_tree_search_insert(struct rmap_item *rmap_item,
1106 struct page *page,
1107 struct page **tree_pagep)
1108
Izik Eidus31dbd012009-09-21 17:02:03 -07001109{
1110 struct rb_node **new = &root_unstable_tree.rb_node;
1111 struct rb_node *parent = NULL;
1112
1113 while (*new) {
1114 struct rmap_item *tree_rmap_item;
Hugh Dickins8dd35572009-12-14 17:59:18 -08001115 struct page *tree_page;
Izik Eidus31dbd012009-09-21 17:02:03 -07001116 int ret;
1117
Hugh Dickinsd178f272009-11-09 15:58:23 +00001118 cond_resched();
Izik Eidus31dbd012009-09-21 17:02:03 -07001119 tree_rmap_item = rb_entry(*new, struct rmap_item, node);
Hugh Dickins8dd35572009-12-14 17:59:18 -08001120 tree_page = get_mergeable_page(tree_rmap_item);
Dan Carpenter22eccdd2010-04-23 13:18:10 -04001121 if (IS_ERR_OR_NULL(tree_page))
Izik Eidus31dbd012009-09-21 17:02:03 -07001122 return NULL;
1123
1124 /*
Hugh Dickins8dd35572009-12-14 17:59:18 -08001125 * Don't substitute a ksm page for a forked page.
Izik Eidus31dbd012009-09-21 17:02:03 -07001126 */
Hugh Dickins8dd35572009-12-14 17:59:18 -08001127 if (page == tree_page) {
1128 put_page(tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001129 return NULL;
1130 }
1131
Hugh Dickins8dd35572009-12-14 17:59:18 -08001132 ret = memcmp_pages(page, tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001133
1134 parent = *new;
1135 if (ret < 0) {
Hugh Dickins8dd35572009-12-14 17:59:18 -08001136 put_page(tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001137 new = &parent->rb_left;
1138 } else if (ret > 0) {
Hugh Dickins8dd35572009-12-14 17:59:18 -08001139 put_page(tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001140 new = &parent->rb_right;
1141 } else {
Hugh Dickins8dd35572009-12-14 17:59:18 -08001142 *tree_pagep = tree_page;
Izik Eidus31dbd012009-09-21 17:02:03 -07001143 return tree_rmap_item;
1144 }
1145 }
1146
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001147 rmap_item->address |= UNSTABLE_FLAG;
Izik Eidus31dbd012009-09-21 17:02:03 -07001148 rmap_item->address |= (ksm_scan.seqnr & SEQNR_MASK);
1149 rb_link_node(&rmap_item->node, parent, new);
1150 rb_insert_color(&rmap_item->node, &root_unstable_tree);
1151
Hugh Dickins473b0ce2009-09-21 17:02:11 -07001152 ksm_pages_unshared++;
Izik Eidus31dbd012009-09-21 17:02:03 -07001153 return NULL;
1154}
1155
1156/*
1157 * stable_tree_append - add another rmap_item to the linked list of
1158 * rmap_items hanging off a given node of the stable tree, all sharing
1159 * the same ksm page.
1160 */
1161static void stable_tree_append(struct rmap_item *rmap_item,
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001162 struct stable_node *stable_node)
Izik Eidus31dbd012009-09-21 17:02:03 -07001163{
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001164 rmap_item->head = stable_node;
Izik Eidus31dbd012009-09-21 17:02:03 -07001165 rmap_item->address |= STABLE_FLAG;
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001166 hlist_add_head(&rmap_item->hlist, &stable_node->hlist);
Hugh Dickinse178dfd2009-09-21 17:02:10 -07001167
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001168 if (rmap_item->hlist.next)
1169 ksm_pages_sharing++;
1170 else
1171 ksm_pages_shared++;
Izik Eidus31dbd012009-09-21 17:02:03 -07001172}
1173
1174/*
Hugh Dickins81464e302009-09-21 17:02:15 -07001175 * cmp_and_merge_page - first see if page can be merged into the stable tree;
1176 * if not, compare checksum to previous and if it's the same, see if page can
1177 * be inserted into the unstable tree, or merged with a page already there and
1178 * both transferred to the stable tree.
Izik Eidus31dbd012009-09-21 17:02:03 -07001179 *
1180 * @page: the page that we are searching identical page to.
1181 * @rmap_item: the reverse mapping into the virtual address of this page
1182 */
1183static void cmp_and_merge_page(struct page *page, struct rmap_item *rmap_item)
1184{
Izik Eidus31dbd012009-09-21 17:02:03 -07001185 struct rmap_item *tree_rmap_item;
Hugh Dickins8dd35572009-12-14 17:59:18 -08001186 struct page *tree_page = NULL;
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001187 struct stable_node *stable_node;
Hugh Dickins8dd35572009-12-14 17:59:18 -08001188 struct page *kpage;
Izik Eidus31dbd012009-09-21 17:02:03 -07001189 unsigned int checksum;
1190 int err;
1191
Hugh Dickins93d17712009-12-14 17:59:16 -08001192 remove_rmap_item_from_tree(rmap_item);
Izik Eidus31dbd012009-09-21 17:02:03 -07001193
1194 /* We first start with searching the page inside the stable tree */
Hugh Dickins62b61f62009-12-14 17:59:33 -08001195 kpage = stable_tree_search(page);
1196 if (kpage) {
Hugh Dickins08beca42009-12-14 17:59:21 -08001197 err = try_to_merge_with_ksm_page(rmap_item, page, kpage);
Izik Eidus31dbd012009-09-21 17:02:03 -07001198 if (!err) {
1199 /*
1200 * The page was successfully merged:
1201 * add its rmap_item to the stable tree.
1202 */
Hugh Dickins5ad64682009-12-14 17:59:24 -08001203 lock_page(kpage);
Hugh Dickins62b61f62009-12-14 17:59:33 -08001204 stable_tree_append(rmap_item, page_stable_node(kpage));
Hugh Dickins5ad64682009-12-14 17:59:24 -08001205 unlock_page(kpage);
Izik Eidus31dbd012009-09-21 17:02:03 -07001206 }
Hugh Dickins8dd35572009-12-14 17:59:18 -08001207 put_page(kpage);
Izik Eidus31dbd012009-09-21 17:02:03 -07001208 return;
1209 }
1210
1211 /*
Hugh Dickins4035c07a2009-12-14 17:59:27 -08001212 * If the hash value of the page has changed from the last time
1213 * we calculated it, this page is changing frequently: therefore we
1214 * don't want to insert it in the unstable tree, and we don't want
1215 * to waste our time searching for something identical to it there.
Izik Eidus31dbd012009-09-21 17:02:03 -07001216 */
1217 checksum = calc_checksum(page);
1218 if (rmap_item->oldchecksum != checksum) {
1219 rmap_item->oldchecksum = checksum;
1220 return;
1221 }
1222
Hugh Dickins8dd35572009-12-14 17:59:18 -08001223 tree_rmap_item =
1224 unstable_tree_search_insert(rmap_item, page, &tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001225 if (tree_rmap_item) {
Hugh Dickins8dd35572009-12-14 17:59:18 -08001226 kpage = try_to_merge_two_pages(rmap_item, page,
1227 tree_rmap_item, tree_page);
1228 put_page(tree_page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001229 /*
1230 * As soon as we merge this page, we want to remove the
1231 * rmap_item of the page we have merged with from the unstable
1232 * tree, and insert it instead as new node in the stable tree.
1233 */
Hugh Dickins8dd35572009-12-14 17:59:18 -08001234 if (kpage) {
Hugh Dickins93d17712009-12-14 17:59:16 -08001235 remove_rmap_item_from_tree(tree_rmap_item);
Hugh Dickins473b0ce2009-09-21 17:02:11 -07001236
Hugh Dickins5ad64682009-12-14 17:59:24 -08001237 lock_page(kpage);
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001238 stable_node = stable_tree_insert(kpage);
1239 if (stable_node) {
1240 stable_tree_append(tree_rmap_item, stable_node);
1241 stable_tree_append(rmap_item, stable_node);
1242 }
Hugh Dickins5ad64682009-12-14 17:59:24 -08001243 unlock_page(kpage);
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001244
Izik Eidus31dbd012009-09-21 17:02:03 -07001245 /*
1246 * If we fail to insert the page into the stable tree,
1247 * we will have 2 virtual addresses that are pointing
1248 * to a ksm page left outside the stable tree,
1249 * in which case we need to break_cow on both.
1250 */
Hugh Dickins7b6ba2c2009-12-14 17:59:20 -08001251 if (!stable_node) {
Hugh Dickins8dd35572009-12-14 17:59:18 -08001252 break_cow(tree_rmap_item);
1253 break_cow(rmap_item);
Izik Eidus31dbd012009-09-21 17:02:03 -07001254 }
1255 }
Izik Eidus31dbd012009-09-21 17:02:03 -07001256 }
1257}
1258
1259static struct rmap_item *get_next_rmap_item(struct mm_slot *mm_slot,
Hugh Dickins6514d512009-12-14 17:59:19 -08001260 struct rmap_item **rmap_list,
Izik Eidus31dbd012009-09-21 17:02:03 -07001261 unsigned long addr)
1262{
1263 struct rmap_item *rmap_item;
1264
Hugh Dickins6514d512009-12-14 17:59:19 -08001265 while (*rmap_list) {
1266 rmap_item = *rmap_list;
Hugh Dickins93d17712009-12-14 17:59:16 -08001267 if ((rmap_item->address & PAGE_MASK) == addr)
Izik Eidus31dbd012009-09-21 17:02:03 -07001268 return rmap_item;
Izik Eidus31dbd012009-09-21 17:02:03 -07001269 if (rmap_item->address > addr)
1270 break;
Hugh Dickins6514d512009-12-14 17:59:19 -08001271 *rmap_list = rmap_item->rmap_list;
Izik Eidus31dbd012009-09-21 17:02:03 -07001272 remove_rmap_item_from_tree(rmap_item);
Izik Eidus31dbd012009-09-21 17:02:03 -07001273 free_rmap_item(rmap_item);
1274 }
1275
1276 rmap_item = alloc_rmap_item();
1277 if (rmap_item) {
1278 /* It has already been zeroed */
1279 rmap_item->mm = mm_slot->mm;
1280 rmap_item->address = addr;
Hugh Dickins6514d512009-12-14 17:59:19 -08001281 rmap_item->rmap_list = *rmap_list;
1282 *rmap_list = rmap_item;
Izik Eidus31dbd012009-09-21 17:02:03 -07001283 }
1284 return rmap_item;
1285}
1286
1287static struct rmap_item *scan_get_next_rmap_item(struct page **page)
1288{
1289 struct mm_struct *mm;
1290 struct mm_slot *slot;
1291 struct vm_area_struct *vma;
1292 struct rmap_item *rmap_item;
1293
1294 if (list_empty(&ksm_mm_head.mm_list))
1295 return NULL;
1296
1297 slot = ksm_scan.mm_slot;
1298 if (slot == &ksm_mm_head) {
Hugh Dickins2919bfd2011-01-13 15:47:29 -08001299 /*
1300 * A number of pages can hang around indefinitely on per-cpu
1301 * pagevecs, raised page count preventing write_protect_page
1302 * from merging them. Though it doesn't really matter much,
1303 * it is puzzling to see some stuck in pages_volatile until
1304 * other activity jostles them out, and they also prevented
1305 * LTP's KSM test from succeeding deterministically; so drain
1306 * them here (here rather than on entry to ksm_do_scan(),
1307 * so we don't IPI too often when pages_to_scan is set low).
1308 */
1309 lru_add_drain_all();
1310
Izik Eidus31dbd012009-09-21 17:02:03 -07001311 root_unstable_tree = RB_ROOT;
1312
1313 spin_lock(&ksm_mmlist_lock);
1314 slot = list_entry(slot->mm_list.next, struct mm_slot, mm_list);
1315 ksm_scan.mm_slot = slot;
1316 spin_unlock(&ksm_mmlist_lock);
1317next_mm:
1318 ksm_scan.address = 0;
Hugh Dickins6514d512009-12-14 17:59:19 -08001319 ksm_scan.rmap_list = &slot->rmap_list;
Izik Eidus31dbd012009-09-21 17:02:03 -07001320 }
1321
1322 mm = slot->mm;
1323 down_read(&mm->mmap_sem);
Hugh Dickins9ba69292009-09-21 17:02:20 -07001324 if (ksm_test_exit(mm))
1325 vma = NULL;
1326 else
1327 vma = find_vma(mm, ksm_scan.address);
1328
1329 for (; vma; vma = vma->vm_next) {
Izik Eidus31dbd012009-09-21 17:02:03 -07001330 if (!(vma->vm_flags & VM_MERGEABLE))
1331 continue;
1332 if (ksm_scan.address < vma->vm_start)
1333 ksm_scan.address = vma->vm_start;
1334 if (!vma->anon_vma)
1335 ksm_scan.address = vma->vm_end;
1336
1337 while (ksm_scan.address < vma->vm_end) {
Hugh Dickins9ba69292009-09-21 17:02:20 -07001338 if (ksm_test_exit(mm))
1339 break;
Izik Eidus31dbd012009-09-21 17:02:03 -07001340 *page = follow_page(vma, ksm_scan.address, FOLL_GET);
Andrea Arcangeli21ae5b02011-01-13 15:47:00 -08001341 if (IS_ERR_OR_NULL(*page)) {
1342 ksm_scan.address += PAGE_SIZE;
1343 cond_resched();
1344 continue;
1345 }
Andrea Arcangeli29ad7682011-01-13 15:47:19 -08001346 if (PageAnon(*page) ||
1347 page_trans_compound_anon(*page)) {
Izik Eidus31dbd012009-09-21 17:02:03 -07001348 flush_anon_page(vma, *page, ksm_scan.address);
1349 flush_dcache_page(*page);
1350 rmap_item = get_next_rmap_item(slot,
Hugh Dickins6514d512009-12-14 17:59:19 -08001351 ksm_scan.rmap_list, ksm_scan.address);
Izik Eidus31dbd012009-09-21 17:02:03 -07001352 if (rmap_item) {
Hugh Dickins6514d512009-12-14 17:59:19 -08001353 ksm_scan.rmap_list =
1354 &rmap_item->rmap_list;
Izik Eidus31dbd012009-09-21 17:02:03 -07001355 ksm_scan.address += PAGE_SIZE;
1356 } else
1357 put_page(*page);
1358 up_read(&mm->mmap_sem);
1359 return rmap_item;
1360 }
Andrea Arcangeli21ae5b02011-01-13 15:47:00 -08001361 put_page(*page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001362 ksm_scan.address += PAGE_SIZE;
1363 cond_resched();
1364 }
1365 }
1366
Hugh Dickins9ba69292009-09-21 17:02:20 -07001367 if (ksm_test_exit(mm)) {
1368 ksm_scan.address = 0;
Hugh Dickins6514d512009-12-14 17:59:19 -08001369 ksm_scan.rmap_list = &slot->rmap_list;
Hugh Dickins9ba69292009-09-21 17:02:20 -07001370 }
Izik Eidus31dbd012009-09-21 17:02:03 -07001371 /*
1372 * Nuke all the rmap_items that are above this current rmap:
1373 * because there were no VM_MERGEABLE vmas with such addresses.
1374 */
Hugh Dickins6514d512009-12-14 17:59:19 -08001375 remove_trailing_rmap_items(slot, ksm_scan.rmap_list);
Izik Eidus31dbd012009-09-21 17:02:03 -07001376
1377 spin_lock(&ksm_mmlist_lock);
Hugh Dickinscd551f92009-09-21 17:02:17 -07001378 ksm_scan.mm_slot = list_entry(slot->mm_list.next,
1379 struct mm_slot, mm_list);
1380 if (ksm_scan.address == 0) {
1381 /*
1382 * We've completed a full scan of all vmas, holding mmap_sem
1383 * throughout, and found no VM_MERGEABLE: so do the same as
1384 * __ksm_exit does to remove this mm from all our lists now.
Hugh Dickins9ba69292009-09-21 17:02:20 -07001385 * This applies either when cleaning up after __ksm_exit
1386 * (but beware: we can reach here even before __ksm_exit),
1387 * or when all VM_MERGEABLE areas have been unmapped (and
1388 * mmap_sem then protects against race with MADV_MERGEABLE).
Hugh Dickinscd551f92009-09-21 17:02:17 -07001389 */
1390 hlist_del(&slot->link);
1391 list_del(&slot->mm_list);
Hugh Dickins9ba69292009-09-21 17:02:20 -07001392 spin_unlock(&ksm_mmlist_lock);
1393
Hugh Dickinscd551f92009-09-21 17:02:17 -07001394 free_mm_slot(slot);
1395 clear_bit(MMF_VM_MERGEABLE, &mm->flags);
Hugh Dickins9ba69292009-09-21 17:02:20 -07001396 up_read(&mm->mmap_sem);
1397 mmdrop(mm);
1398 } else {
1399 spin_unlock(&ksm_mmlist_lock);
1400 up_read(&mm->mmap_sem);
Hugh Dickinscd551f92009-09-21 17:02:17 -07001401 }
Izik Eidus31dbd012009-09-21 17:02:03 -07001402
1403 /* Repeat until we've completed scanning the whole list */
Hugh Dickinscd551f92009-09-21 17:02:17 -07001404 slot = ksm_scan.mm_slot;
Izik Eidus31dbd012009-09-21 17:02:03 -07001405 if (slot != &ksm_mm_head)
1406 goto next_mm;
1407
Izik Eidus31dbd012009-09-21 17:02:03 -07001408 ksm_scan.seqnr++;
1409 return NULL;
1410}
1411
1412/**
1413 * ksm_do_scan - the ksm scanner main worker function.
1414 * @scan_npages - number of pages we want to scan before we return.
1415 */
1416static void ksm_do_scan(unsigned int scan_npages)
1417{
1418 struct rmap_item *rmap_item;
Dan Carpenter22eccdd2010-04-23 13:18:10 -04001419 struct page *uninitialized_var(page);
Izik Eidus31dbd012009-09-21 17:02:03 -07001420
Andrea Arcangeli878aee72011-01-13 15:47:10 -08001421 while (scan_npages-- && likely(!freezing(current))) {
Izik Eidus31dbd012009-09-21 17:02:03 -07001422 cond_resched();
1423 rmap_item = scan_get_next_rmap_item(&page);
1424 if (!rmap_item)
1425 return;
1426 if (!PageKsm(page) || !in_stable_tree(rmap_item))
1427 cmp_and_merge_page(page, rmap_item);
1428 put_page(page);
1429 }
1430}
1431
Hugh Dickins6e1583842009-09-21 17:02:14 -07001432static int ksmd_should_run(void)
1433{
1434 return (ksm_run & KSM_RUN_MERGE) && !list_empty(&ksm_mm_head.mm_list);
1435}
1436
Izik Eidus31dbd012009-09-21 17:02:03 -07001437static int ksm_scan_thread(void *nothing)
1438{
Andrea Arcangeli878aee72011-01-13 15:47:10 -08001439 set_freezable();
Izik Eidus339aa622009-09-21 17:02:07 -07001440 set_user_nice(current, 5);
Izik Eidus31dbd012009-09-21 17:02:03 -07001441
1442 while (!kthread_should_stop()) {
Hugh Dickins6e1583842009-09-21 17:02:14 -07001443 mutex_lock(&ksm_thread_mutex);
1444 if (ksmd_should_run())
Izik Eidus31dbd012009-09-21 17:02:03 -07001445 ksm_do_scan(ksm_thread_pages_to_scan);
Hugh Dickins6e1583842009-09-21 17:02:14 -07001446 mutex_unlock(&ksm_thread_mutex);
1447
Andrea Arcangeli878aee72011-01-13 15:47:10 -08001448 try_to_freeze();
1449
Hugh Dickins6e1583842009-09-21 17:02:14 -07001450 if (ksmd_should_run()) {
Izik Eidus31dbd012009-09-21 17:02:03 -07001451 schedule_timeout_interruptible(
1452 msecs_to_jiffies(ksm_thread_sleep_millisecs));
1453 } else {
Andrea Arcangeli878aee72011-01-13 15:47:10 -08001454 wait_event_freezable(ksm_thread_wait,
Hugh Dickins6e1583842009-09-21 17:02:14 -07001455 ksmd_should_run() || kthread_should_stop());
Izik Eidus31dbd012009-09-21 17:02:03 -07001456 }
1457 }
1458 return 0;
1459}
1460
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001461int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
1462 unsigned long end, int advice, unsigned long *vm_flags)
1463{
1464 struct mm_struct *mm = vma->vm_mm;
Hugh Dickinsd952b792009-09-21 17:02:16 -07001465 int err;
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001466
1467 switch (advice) {
1468 case MADV_MERGEABLE:
1469 /*
1470 * Be somewhat over-protective for now!
1471 */
1472 if (*vm_flags & (VM_MERGEABLE | VM_SHARED | VM_MAYSHARE |
1473 VM_PFNMAP | VM_IO | VM_DONTEXPAND |
1474 VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE |
Hugh Dickins5ad64682009-12-14 17:59:24 -08001475 VM_NONLINEAR | VM_MIXEDMAP | VM_SAO))
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001476 return 0; /* just ignore the advice */
1477
Hugh Dickinsd952b792009-09-21 17:02:16 -07001478 if (!test_bit(MMF_VM_MERGEABLE, &mm->flags)) {
1479 err = __ksm_enter(mm);
1480 if (err)
1481 return err;
1482 }
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001483
1484 *vm_flags |= VM_MERGEABLE;
1485 break;
1486
1487 case MADV_UNMERGEABLE:
1488 if (!(*vm_flags & VM_MERGEABLE))
1489 return 0; /* just ignore the advice */
1490
Hugh Dickinsd952b792009-09-21 17:02:16 -07001491 if (vma->anon_vma) {
1492 err = unmerge_ksm_pages(vma, start, end);
1493 if (err)
1494 return err;
1495 }
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001496
1497 *vm_flags &= ~VM_MERGEABLE;
1498 break;
1499 }
1500
1501 return 0;
1502}
1503
1504int __ksm_enter(struct mm_struct *mm)
1505{
Hugh Dickins6e1583842009-09-21 17:02:14 -07001506 struct mm_slot *mm_slot;
1507 int needs_wakeup;
1508
1509 mm_slot = alloc_mm_slot();
Izik Eidus31dbd012009-09-21 17:02:03 -07001510 if (!mm_slot)
1511 return -ENOMEM;
1512
Hugh Dickins6e1583842009-09-21 17:02:14 -07001513 /* Check ksm_run too? Would need tighter locking */
1514 needs_wakeup = list_empty(&ksm_mm_head.mm_list);
1515
Izik Eidus31dbd012009-09-21 17:02:03 -07001516 spin_lock(&ksm_mmlist_lock);
1517 insert_to_mm_slots_hash(mm, mm_slot);
1518 /*
1519 * Insert just behind the scanning cursor, to let the area settle
1520 * down a little; when fork is followed by immediate exec, we don't
1521 * want ksmd to waste time setting up and tearing down an rmap_list.
1522 */
1523 list_add_tail(&mm_slot->mm_list, &ksm_scan.mm_slot->mm_list);
1524 spin_unlock(&ksm_mmlist_lock);
1525
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001526 set_bit(MMF_VM_MERGEABLE, &mm->flags);
Hugh Dickins9ba69292009-09-21 17:02:20 -07001527 atomic_inc(&mm->mm_count);
Hugh Dickins6e1583842009-09-21 17:02:14 -07001528
1529 if (needs_wakeup)
1530 wake_up_interruptible(&ksm_thread_wait);
1531
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001532 return 0;
1533}
1534
Andrea Arcangeli1c2fb7a2009-09-21 17:02:22 -07001535void __ksm_exit(struct mm_struct *mm)
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001536{
Hugh Dickinscd551f92009-09-21 17:02:17 -07001537 struct mm_slot *mm_slot;
Hugh Dickins9ba69292009-09-21 17:02:20 -07001538 int easy_to_free = 0;
Hugh Dickinscd551f92009-09-21 17:02:17 -07001539
Izik Eidus31dbd012009-09-21 17:02:03 -07001540 /*
Hugh Dickins9ba69292009-09-21 17:02:20 -07001541 * This process is exiting: if it's straightforward (as is the
1542 * case when ksmd was never running), free mm_slot immediately.
1543 * But if it's at the cursor or has rmap_items linked to it, use
1544 * mmap_sem to synchronize with any break_cows before pagetables
1545 * are freed, and leave the mm_slot on the list for ksmd to free.
1546 * Beware: ksm may already have noticed it exiting and freed the slot.
Izik Eidus31dbd012009-09-21 17:02:03 -07001547 */
Hugh Dickins9ba69292009-09-21 17:02:20 -07001548
Hugh Dickinscd551f92009-09-21 17:02:17 -07001549 spin_lock(&ksm_mmlist_lock);
1550 mm_slot = get_mm_slot(mm);
Hugh Dickins9ba69292009-09-21 17:02:20 -07001551 if (mm_slot && ksm_scan.mm_slot != mm_slot) {
Hugh Dickins6514d512009-12-14 17:59:19 -08001552 if (!mm_slot->rmap_list) {
Hugh Dickins9ba69292009-09-21 17:02:20 -07001553 hlist_del(&mm_slot->link);
1554 list_del(&mm_slot->mm_list);
1555 easy_to_free = 1;
1556 } else {
1557 list_move(&mm_slot->mm_list,
1558 &ksm_scan.mm_slot->mm_list);
1559 }
Hugh Dickinscd551f92009-09-21 17:02:17 -07001560 }
Hugh Dickinscd551f92009-09-21 17:02:17 -07001561 spin_unlock(&ksm_mmlist_lock);
1562
Hugh Dickins9ba69292009-09-21 17:02:20 -07001563 if (easy_to_free) {
1564 free_mm_slot(mm_slot);
1565 clear_bit(MMF_VM_MERGEABLE, &mm->flags);
1566 mmdrop(mm);
1567 } else if (mm_slot) {
Hugh Dickins9ba69292009-09-21 17:02:20 -07001568 down_write(&mm->mmap_sem);
1569 up_write(&mm->mmap_sem);
Hugh Dickins9ba69292009-09-21 17:02:20 -07001570 }
Hugh Dickinsf8af4da2009-09-21 17:01:57 -07001571}
Izik Eidus31dbd012009-09-21 17:02:03 -07001572
Hugh Dickins5ad64682009-12-14 17:59:24 -08001573struct page *ksm_does_need_to_copy(struct page *page,
1574 struct vm_area_struct *vma, unsigned long address)
1575{
1576 struct page *new_page;
1577
Hugh Dickins5ad64682009-12-14 17:59:24 -08001578 new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address);
1579 if (new_page) {
1580 copy_user_highpage(new_page, page, address, vma);
1581
1582 SetPageDirty(new_page);
1583 __SetPageUptodate(new_page);
1584 SetPageSwapBacked(new_page);
1585 __set_page_locked(new_page);
1586
1587 if (page_evictable(new_page, vma))
1588 lru_cache_add_lru(new_page, LRU_ACTIVE_ANON);
1589 else
1590 add_page_to_unevictable_list(new_page);
1591 }
1592
Hugh Dickins5ad64682009-12-14 17:59:24 -08001593 return new_page;
1594}
1595
1596int page_referenced_ksm(struct page *page, struct mem_cgroup *memcg,
1597 unsigned long *vm_flags)
1598{
1599 struct stable_node *stable_node;
1600 struct rmap_item *rmap_item;
1601 struct hlist_node *hlist;
1602 unsigned int mapcount = page_mapcount(page);
1603 int referenced = 0;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001604 int search_new_forks = 0;
Hugh Dickins5ad64682009-12-14 17:59:24 -08001605
1606 VM_BUG_ON(!PageKsm(page));
1607 VM_BUG_ON(!PageLocked(page));
1608
1609 stable_node = page_stable_node(page);
1610 if (!stable_node)
1611 return 0;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001612again:
Hugh Dickins5ad64682009-12-14 17:59:24 -08001613 hlist_for_each_entry(rmap_item, hlist, &stable_node->hlist, hlist) {
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001614 struct anon_vma *anon_vma = rmap_item->anon_vma;
Rik van Riel5beb4932010-03-05 13:42:07 -08001615 struct anon_vma_chain *vmac;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001616 struct vm_area_struct *vma;
Hugh Dickins5ad64682009-12-14 17:59:24 -08001617
Rik van Rielcba48b92010-08-09 17:18:38 -07001618 anon_vma_lock(anon_vma);
Rik van Riel5beb4932010-03-05 13:42:07 -08001619 list_for_each_entry(vmac, &anon_vma->head, same_anon_vma) {
1620 vma = vmac->vma;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001621 if (rmap_item->address < vma->vm_start ||
1622 rmap_item->address >= vma->vm_end)
1623 continue;
1624 /*
1625 * Initially we examine only the vma which covers this
1626 * rmap_item; but later, if there is still work to do,
1627 * we examine covering vmas in other mms: in case they
1628 * were forked from the original since ksmd passed.
1629 */
1630 if ((rmap_item->mm == vma->vm_mm) == search_new_forks)
1631 continue;
Hugh Dickins5ad64682009-12-14 17:59:24 -08001632
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001633 if (memcg && !mm_match_cgroup(vma->vm_mm, memcg))
1634 continue;
1635
1636 referenced += page_referenced_one(page, vma,
Hugh Dickins5ad64682009-12-14 17:59:24 -08001637 rmap_item->address, &mapcount, vm_flags);
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001638 if (!search_new_forks || !mapcount)
1639 break;
1640 }
Rik van Rielcba48b92010-08-09 17:18:38 -07001641 anon_vma_unlock(anon_vma);
Hugh Dickins5ad64682009-12-14 17:59:24 -08001642 if (!mapcount)
1643 goto out;
1644 }
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001645 if (!search_new_forks++)
1646 goto again;
Hugh Dickins5ad64682009-12-14 17:59:24 -08001647out:
Hugh Dickins5ad64682009-12-14 17:59:24 -08001648 return referenced;
1649}
1650
1651int try_to_unmap_ksm(struct page *page, enum ttu_flags flags)
1652{
1653 struct stable_node *stable_node;
1654 struct hlist_node *hlist;
1655 struct rmap_item *rmap_item;
1656 int ret = SWAP_AGAIN;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001657 int search_new_forks = 0;
Hugh Dickins5ad64682009-12-14 17:59:24 -08001658
1659 VM_BUG_ON(!PageKsm(page));
1660 VM_BUG_ON(!PageLocked(page));
1661
1662 stable_node = page_stable_node(page);
1663 if (!stable_node)
1664 return SWAP_FAIL;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001665again:
Hugh Dickins5ad64682009-12-14 17:59:24 -08001666 hlist_for_each_entry(rmap_item, hlist, &stable_node->hlist, hlist) {
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001667 struct anon_vma *anon_vma = rmap_item->anon_vma;
Rik van Riel5beb4932010-03-05 13:42:07 -08001668 struct anon_vma_chain *vmac;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001669 struct vm_area_struct *vma;
Hugh Dickins5ad64682009-12-14 17:59:24 -08001670
Rik van Rielcba48b92010-08-09 17:18:38 -07001671 anon_vma_lock(anon_vma);
Rik van Riel5beb4932010-03-05 13:42:07 -08001672 list_for_each_entry(vmac, &anon_vma->head, same_anon_vma) {
1673 vma = vmac->vma;
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001674 if (rmap_item->address < vma->vm_start ||
1675 rmap_item->address >= vma->vm_end)
1676 continue;
1677 /*
1678 * Initially we examine only the vma which covers this
1679 * rmap_item; but later, if there is still work to do,
1680 * we examine covering vmas in other mms: in case they
1681 * were forked from the original since ksmd passed.
1682 */
1683 if ((rmap_item->mm == vma->vm_mm) == search_new_forks)
1684 continue;
1685
1686 ret = try_to_unmap_one(page, vma,
1687 rmap_item->address, flags);
1688 if (ret != SWAP_AGAIN || !page_mapped(page)) {
Rik van Rielcba48b92010-08-09 17:18:38 -07001689 anon_vma_unlock(anon_vma);
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001690 goto out;
1691 }
1692 }
Rik van Rielcba48b92010-08-09 17:18:38 -07001693 anon_vma_unlock(anon_vma);
Hugh Dickins5ad64682009-12-14 17:59:24 -08001694 }
Hugh Dickinsdb114b82009-12-14 17:59:25 -08001695 if (!search_new_forks++)
1696 goto again;
Hugh Dickins5ad64682009-12-14 17:59:24 -08001697out:
Hugh Dickins5ad64682009-12-14 17:59:24 -08001698 return ret;
1699}
1700
Hugh Dickinse9995ef2009-12-14 17:59:31 -08001701#ifdef CONFIG_MIGRATION
1702int rmap_walk_ksm(struct page *page, int (*rmap_one)(struct page *,
1703 struct vm_area_struct *, unsigned long, void *), void *arg)
1704{
1705 struct stable_node *stable_node;
1706 struct hlist_node *hlist;
1707 struct rmap_item *rmap_item;
1708 int ret = SWAP_AGAIN;
1709 int search_new_forks = 0;
1710
1711 VM_BUG_ON(!PageKsm(page));
1712 VM_BUG_ON(!PageLocked(page));
1713
1714 stable_node = page_stable_node(page);
1715 if (!stable_node)
1716 return ret;
1717again:
1718 hlist_for_each_entry(rmap_item, hlist, &stable_node->hlist, hlist) {
1719 struct anon_vma *anon_vma = rmap_item->anon_vma;
Rik van Riel5beb4932010-03-05 13:42:07 -08001720 struct anon_vma_chain *vmac;
Hugh Dickinse9995ef2009-12-14 17:59:31 -08001721 struct vm_area_struct *vma;
1722
Rik van Rielcba48b92010-08-09 17:18:38 -07001723 anon_vma_lock(anon_vma);
Rik van Riel5beb4932010-03-05 13:42:07 -08001724 list_for_each_entry(vmac, &anon_vma->head, same_anon_vma) {
1725 vma = vmac->vma;
Hugh Dickinse9995ef2009-12-14 17:59:31 -08001726 if (rmap_item->address < vma->vm_start ||
1727 rmap_item->address >= vma->vm_end)
1728 continue;
1729 /*
1730 * Initially we examine only the vma which covers this
1731 * rmap_item; but later, if there is still work to do,
1732 * we examine covering vmas in other mms: in case they
1733 * were forked from the original since ksmd passed.
1734 */
1735 if ((rmap_item->mm == vma->vm_mm) == search_new_forks)
1736 continue;
1737
1738 ret = rmap_one(page, vma, rmap_item->address, arg);
1739 if (ret != SWAP_AGAIN) {
Rik van Rielcba48b92010-08-09 17:18:38 -07001740 anon_vma_unlock(anon_vma);
Hugh Dickinse9995ef2009-12-14 17:59:31 -08001741 goto out;
1742 }
1743 }
Rik van Rielcba48b92010-08-09 17:18:38 -07001744 anon_vma_unlock(anon_vma);
Hugh Dickinse9995ef2009-12-14 17:59:31 -08001745 }
1746 if (!search_new_forks++)
1747 goto again;
1748out:
1749 return ret;
1750}
1751
1752void ksm_migrate_page(struct page *newpage, struct page *oldpage)
1753{
1754 struct stable_node *stable_node;
1755
1756 VM_BUG_ON(!PageLocked(oldpage));
1757 VM_BUG_ON(!PageLocked(newpage));
1758 VM_BUG_ON(newpage->mapping != oldpage->mapping);
1759
1760 stable_node = page_stable_node(newpage);
1761 if (stable_node) {
Hugh Dickins62b61f62009-12-14 17:59:33 -08001762 VM_BUG_ON(stable_node->kpfn != page_to_pfn(oldpage));
1763 stable_node->kpfn = page_to_pfn(newpage);
Hugh Dickinse9995ef2009-12-14 17:59:31 -08001764 }
1765}
1766#endif /* CONFIG_MIGRATION */
1767
Hugh Dickins62b61f62009-12-14 17:59:33 -08001768#ifdef CONFIG_MEMORY_HOTREMOVE
1769static struct stable_node *ksm_check_stable_tree(unsigned long start_pfn,
1770 unsigned long end_pfn)
1771{
1772 struct rb_node *node;
1773
1774 for (node = rb_first(&root_stable_tree); node; node = rb_next(node)) {
1775 struct stable_node *stable_node;
1776
1777 stable_node = rb_entry(node, struct stable_node, node);
1778 if (stable_node->kpfn >= start_pfn &&
1779 stable_node->kpfn < end_pfn)
1780 return stable_node;
1781 }
1782 return NULL;
1783}
1784
1785static int ksm_memory_callback(struct notifier_block *self,
1786 unsigned long action, void *arg)
1787{
1788 struct memory_notify *mn = arg;
1789 struct stable_node *stable_node;
1790
1791 switch (action) {
1792 case MEM_GOING_OFFLINE:
1793 /*
1794 * Keep it very simple for now: just lock out ksmd and
1795 * MADV_UNMERGEABLE while any memory is going offline.
KOSAKI Motohiroa0b0f582010-12-02 14:31:20 -08001796 * mutex_lock_nested() is necessary because lockdep was alarmed
1797 * that here we take ksm_thread_mutex inside notifier chain
1798 * mutex, and later take notifier chain mutex inside
1799 * ksm_thread_mutex to unlock it. But that's safe because both
1800 * are inside mem_hotplug_mutex.
Hugh Dickins62b61f62009-12-14 17:59:33 -08001801 */
KOSAKI Motohiroa0b0f582010-12-02 14:31:20 -08001802 mutex_lock_nested(&ksm_thread_mutex, SINGLE_DEPTH_NESTING);
Hugh Dickins62b61f62009-12-14 17:59:33 -08001803 break;
1804
1805 case MEM_OFFLINE:
1806 /*
1807 * Most of the work is done by page migration; but there might
1808 * be a few stable_nodes left over, still pointing to struct
1809 * pages which have been offlined: prune those from the tree.
1810 */
1811 while ((stable_node = ksm_check_stable_tree(mn->start_pfn,
1812 mn->start_pfn + mn->nr_pages)) != NULL)
1813 remove_node_from_stable_tree(stable_node);
1814 /* fallthrough */
1815
1816 case MEM_CANCEL_OFFLINE:
1817 mutex_unlock(&ksm_thread_mutex);
1818 break;
1819 }
1820 return NOTIFY_OK;
1821}
1822#endif /* CONFIG_MEMORY_HOTREMOVE */
1823
Hugh Dickins2ffd8672009-09-21 17:02:23 -07001824#ifdef CONFIG_SYSFS
1825/*
1826 * This all compiles without CONFIG_SYSFS, but is a waste of space.
1827 */
1828
Izik Eidus31dbd012009-09-21 17:02:03 -07001829#define KSM_ATTR_RO(_name) \
1830 static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
1831#define KSM_ATTR(_name) \
1832 static struct kobj_attribute _name##_attr = \
1833 __ATTR(_name, 0644, _name##_show, _name##_store)
1834
1835static ssize_t sleep_millisecs_show(struct kobject *kobj,
1836 struct kobj_attribute *attr, char *buf)
1837{
1838 return sprintf(buf, "%u\n", ksm_thread_sleep_millisecs);
1839}
1840
1841static ssize_t sleep_millisecs_store(struct kobject *kobj,
1842 struct kobj_attribute *attr,
1843 const char *buf, size_t count)
1844{
1845 unsigned long msecs;
1846 int err;
1847
1848 err = strict_strtoul(buf, 10, &msecs);
1849 if (err || msecs > UINT_MAX)
1850 return -EINVAL;
1851
1852 ksm_thread_sleep_millisecs = msecs;
1853
1854 return count;
1855}
1856KSM_ATTR(sleep_millisecs);
1857
1858static ssize_t pages_to_scan_show(struct kobject *kobj,
1859 struct kobj_attribute *attr, char *buf)
1860{
1861 return sprintf(buf, "%u\n", ksm_thread_pages_to_scan);
1862}
1863
1864static ssize_t pages_to_scan_store(struct kobject *kobj,
1865 struct kobj_attribute *attr,
1866 const char *buf, size_t count)
1867{
1868 int err;
1869 unsigned long nr_pages;
1870
1871 err = strict_strtoul(buf, 10, &nr_pages);
1872 if (err || nr_pages > UINT_MAX)
1873 return -EINVAL;
1874
1875 ksm_thread_pages_to_scan = nr_pages;
1876
1877 return count;
1878}
1879KSM_ATTR(pages_to_scan);
1880
1881static ssize_t run_show(struct kobject *kobj, struct kobj_attribute *attr,
1882 char *buf)
1883{
1884 return sprintf(buf, "%u\n", ksm_run);
1885}
1886
1887static ssize_t run_store(struct kobject *kobj, struct kobj_attribute *attr,
1888 const char *buf, size_t count)
1889{
1890 int err;
1891 unsigned long flags;
1892
1893 err = strict_strtoul(buf, 10, &flags);
1894 if (err || flags > UINT_MAX)
1895 return -EINVAL;
1896 if (flags > KSM_RUN_UNMERGE)
1897 return -EINVAL;
1898
1899 /*
1900 * KSM_RUN_MERGE sets ksmd running, and 0 stops it running.
1901 * KSM_RUN_UNMERGE stops it running and unmerges all rmap_items,
Hugh Dickinsd0f209f2009-12-14 17:59:34 -08001902 * breaking COW to free the pages_shared (but leaves mm_slots
1903 * on the list for when ksmd may be set running again).
Izik Eidus31dbd012009-09-21 17:02:03 -07001904 */
1905
1906 mutex_lock(&ksm_thread_mutex);
1907 if (ksm_run != flags) {
1908 ksm_run = flags;
Hugh Dickinsd952b792009-09-21 17:02:16 -07001909 if (flags & KSM_RUN_UNMERGE) {
Hugh Dickins35451be2009-09-21 17:02:27 -07001910 current->flags |= PF_OOM_ORIGIN;
Hugh Dickinsd952b792009-09-21 17:02:16 -07001911 err = unmerge_and_remove_all_rmap_items();
Hugh Dickins35451be2009-09-21 17:02:27 -07001912 current->flags &= ~PF_OOM_ORIGIN;
Hugh Dickinsd952b792009-09-21 17:02:16 -07001913 if (err) {
1914 ksm_run = KSM_RUN_STOP;
1915 count = err;
1916 }
1917 }
Izik Eidus31dbd012009-09-21 17:02:03 -07001918 }
1919 mutex_unlock(&ksm_thread_mutex);
1920
1921 if (flags & KSM_RUN_MERGE)
1922 wake_up_interruptible(&ksm_thread_wait);
1923
1924 return count;
1925}
1926KSM_ATTR(run);
1927
Hugh Dickinsb4028262009-09-21 17:02:09 -07001928static ssize_t pages_shared_show(struct kobject *kobj,
1929 struct kobj_attribute *attr, char *buf)
1930{
1931 return sprintf(buf, "%lu\n", ksm_pages_shared);
1932}
1933KSM_ATTR_RO(pages_shared);
1934
1935static ssize_t pages_sharing_show(struct kobject *kobj,
1936 struct kobj_attribute *attr, char *buf)
1937{
Hugh Dickinse178dfd2009-09-21 17:02:10 -07001938 return sprintf(buf, "%lu\n", ksm_pages_sharing);
Hugh Dickinsb4028262009-09-21 17:02:09 -07001939}
1940KSM_ATTR_RO(pages_sharing);
1941
Hugh Dickins473b0ce2009-09-21 17:02:11 -07001942static ssize_t pages_unshared_show(struct kobject *kobj,
1943 struct kobj_attribute *attr, char *buf)
1944{
1945 return sprintf(buf, "%lu\n", ksm_pages_unshared);
1946}
1947KSM_ATTR_RO(pages_unshared);
1948
1949static ssize_t pages_volatile_show(struct kobject *kobj,
1950 struct kobj_attribute *attr, char *buf)
1951{
1952 long ksm_pages_volatile;
1953
1954 ksm_pages_volatile = ksm_rmap_items - ksm_pages_shared
1955 - ksm_pages_sharing - ksm_pages_unshared;
1956 /*
1957 * It was not worth any locking to calculate that statistic,
1958 * but it might therefore sometimes be negative: conceal that.
1959 */
1960 if (ksm_pages_volatile < 0)
1961 ksm_pages_volatile = 0;
1962 return sprintf(buf, "%ld\n", ksm_pages_volatile);
1963}
1964KSM_ATTR_RO(pages_volatile);
1965
1966static ssize_t full_scans_show(struct kobject *kobj,
1967 struct kobj_attribute *attr, char *buf)
1968{
1969 return sprintf(buf, "%lu\n", ksm_scan.seqnr);
1970}
1971KSM_ATTR_RO(full_scans);
1972
Izik Eidus31dbd012009-09-21 17:02:03 -07001973static struct attribute *ksm_attrs[] = {
1974 &sleep_millisecs_attr.attr,
1975 &pages_to_scan_attr.attr,
1976 &run_attr.attr,
Hugh Dickinsb4028262009-09-21 17:02:09 -07001977 &pages_shared_attr.attr,
1978 &pages_sharing_attr.attr,
Hugh Dickins473b0ce2009-09-21 17:02:11 -07001979 &pages_unshared_attr.attr,
1980 &pages_volatile_attr.attr,
1981 &full_scans_attr.attr,
Izik Eidus31dbd012009-09-21 17:02:03 -07001982 NULL,
1983};
1984
1985static struct attribute_group ksm_attr_group = {
1986 .attrs = ksm_attrs,
1987 .name = "ksm",
1988};
Hugh Dickins2ffd8672009-09-21 17:02:23 -07001989#endif /* CONFIG_SYSFS */
Izik Eidus31dbd012009-09-21 17:02:03 -07001990
1991static int __init ksm_init(void)
1992{
1993 struct task_struct *ksm_thread;
1994 int err;
1995
1996 err = ksm_slab_init();
1997 if (err)
1998 goto out;
1999
Izik Eidus31dbd012009-09-21 17:02:03 -07002000 ksm_thread = kthread_run(ksm_scan_thread, NULL, "ksmd");
2001 if (IS_ERR(ksm_thread)) {
2002 printk(KERN_ERR "ksm: creating kthread failed\n");
2003 err = PTR_ERR(ksm_thread);
Lai Jiangshand9f89842010-08-09 17:20:02 -07002004 goto out_free;
Izik Eidus31dbd012009-09-21 17:02:03 -07002005 }
2006
Hugh Dickins2ffd8672009-09-21 17:02:23 -07002007#ifdef CONFIG_SYSFS
Izik Eidus31dbd012009-09-21 17:02:03 -07002008 err = sysfs_create_group(mm_kobj, &ksm_attr_group);
2009 if (err) {
2010 printk(KERN_ERR "ksm: register sysfs failed\n");
Hugh Dickins2ffd8672009-09-21 17:02:23 -07002011 kthread_stop(ksm_thread);
Lai Jiangshand9f89842010-08-09 17:20:02 -07002012 goto out_free;
Izik Eidus31dbd012009-09-21 17:02:03 -07002013 }
Hugh Dickinsc73602a2009-10-07 16:32:22 -07002014#else
2015 ksm_run = KSM_RUN_MERGE; /* no way for user to start it */
2016
Hugh Dickins2ffd8672009-09-21 17:02:23 -07002017#endif /* CONFIG_SYSFS */
Izik Eidus31dbd012009-09-21 17:02:03 -07002018
Hugh Dickins62b61f62009-12-14 17:59:33 -08002019#ifdef CONFIG_MEMORY_HOTREMOVE
2020 /*
2021 * Choose a high priority since the callback takes ksm_thread_mutex:
2022 * later callbacks could only be taking locks which nest within that.
2023 */
2024 hotplug_memory_notifier(ksm_memory_callback, 100);
2025#endif
Izik Eidus31dbd012009-09-21 17:02:03 -07002026 return 0;
2027
Lai Jiangshand9f89842010-08-09 17:20:02 -07002028out_free:
Izik Eidus31dbd012009-09-21 17:02:03 -07002029 ksm_slab_free();
2030out:
2031 return err;
2032}
2033module_init(ksm_init)