Blame view
mm/khugepaged.c
54.6 KB
b24413180 License cleanup: ... |
1 |
// SPDX-License-Identifier: GPL-2.0 |
b46e756f5 thp: extract khug... |
2 3 4 5 |
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/mm.h> #include <linux/sched.h> |
6e84f3152 sched/headers: Pr... |
6 |
#include <linux/sched/mm.h> |
f7ccbae45 sched/headers: Pr... |
7 |
#include <linux/sched/coredump.h> |
b46e756f5 thp: extract khug... |
8 9 10 11 12 13 14 15 16 17 18 19 |
#include <linux/mmu_notifier.h> #include <linux/rmap.h> #include <linux/swap.h> #include <linux/mm_inline.h> #include <linux/kthread.h> #include <linux/khugepaged.h> #include <linux/freezer.h> #include <linux/mman.h> #include <linux/hashtable.h> #include <linux/userfaultfd_k.h> #include <linux/page_idle.h> #include <linux/swapops.h> |
f3f0e1d21 khugepaged: add s... |
20 |
#include <linux/shmem_fs.h> |
b46e756f5 thp: extract khug... |
21 22 23 24 25 26 27 28 29 30 31 32 |
#include <asm/tlb.h> #include <asm/pgalloc.h> #include "internal.h" enum scan_result { SCAN_FAIL, SCAN_SUCCEED, SCAN_PMD_NULL, SCAN_EXCEED_NONE_PTE, SCAN_PTE_NON_PRESENT, SCAN_PAGE_RO, |
0db501f7a mm, thp: convert ... |
33 |
SCAN_LACK_REFERENCED_PAGE, |
b46e756f5 thp: extract khug... |
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 |
SCAN_PAGE_NULL, SCAN_SCAN_ABORT, SCAN_PAGE_COUNT, SCAN_PAGE_LRU, SCAN_PAGE_LOCK, SCAN_PAGE_ANON, SCAN_PAGE_COMPOUND, SCAN_ANY_PROCESS, SCAN_VMA_NULL, SCAN_VMA_CHECK, SCAN_ADDRESS_RANGE, SCAN_SWAP_CACHE_PAGE, SCAN_DEL_PAGE_LRU, SCAN_ALLOC_HUGE_PAGE_FAIL, SCAN_CGROUP_CHARGE_FAIL, |
f3f0e1d21 khugepaged: add s... |
49 50 |
SCAN_EXCEED_SWAP_PTE, SCAN_TRUNCATED, |
99cb0dbd4 mm,thp: add read-... |
51 |
SCAN_PAGE_HAS_PRIVATE, |
b46e756f5 thp: extract khug... |
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 |
}; #define CREATE_TRACE_POINTS #include <trace/events/huge_memory.h> /* default scan 8*512 pte (or vmas) every 30 second */ static unsigned int khugepaged_pages_to_scan __read_mostly; static unsigned int khugepaged_pages_collapsed; static unsigned int khugepaged_full_scans; static unsigned int khugepaged_scan_sleep_millisecs __read_mostly = 10000; /* during fragmentation poll the hugepage allocator once every minute */ static unsigned int khugepaged_alloc_sleep_millisecs __read_mostly = 60000; static unsigned long khugepaged_sleep_expire; static DEFINE_SPINLOCK(khugepaged_mm_lock); static DECLARE_WAIT_QUEUE_HEAD(khugepaged_wait); /* * default collapse hugepages if there is at least one pte mapped like * it would have happened if the vma was large enough during page * fault. */ static unsigned int khugepaged_max_ptes_none __read_mostly; static unsigned int khugepaged_max_ptes_swap __read_mostly; #define MM_SLOTS_HASH_BITS 10 static __read_mostly DEFINE_HASHTABLE(mm_slots_hash, MM_SLOTS_HASH_BITS); static struct kmem_cache *mm_slot_cache __read_mostly; |
27e1f8273 khugepaged: enabl... |
79 |
#define MAX_PTE_MAPPED_THP 8 |
b46e756f5 thp: extract khug... |
80 81 82 83 84 85 86 87 88 89 |
/** * struct mm_slot - hash lookup from mm to mm_slot * @hash: hash collision list * @mm_node: khugepaged scan list headed in khugepaged_scan.mm_head * @mm: the mm that this information is valid for */ struct mm_slot { struct hlist_node hash; struct list_head mm_node; struct mm_struct *mm; |
27e1f8273 khugepaged: enabl... |
90 91 92 93 |
/* pte-mapped THP in this mm */ int nr_pte_mapped_thp; unsigned long pte_mapped_thp[MAX_PTE_MAPPED_THP]; |
b46e756f5 thp: extract khug... |
94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 |
}; /** * struct khugepaged_scan - cursor for scanning * @mm_head: the head of the mm list to scan * @mm_slot: the current mm_slot we are scanning * @address: the next address inside that to be scanned * * There is only the one khugepaged_scan instance of this cursor structure. */ struct khugepaged_scan { struct list_head mm_head; struct mm_slot *mm_slot; unsigned long address; }; static struct khugepaged_scan khugepaged_scan = { .mm_head = LIST_HEAD_INIT(khugepaged_scan.mm_head), }; |
e1465d125 mm, thp: propagat... |
113 |
#ifdef CONFIG_SYSFS |
b46e756f5 thp: extract khug... |
114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 |
static ssize_t scan_sleep_millisecs_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return sprintf(buf, "%u ", khugepaged_scan_sleep_millisecs); } static ssize_t scan_sleep_millisecs_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t count) { unsigned long msecs; int err; err = kstrtoul(buf, 10, &msecs); if (err || msecs > UINT_MAX) return -EINVAL; khugepaged_scan_sleep_millisecs = msecs; khugepaged_sleep_expire = 0; wake_up_interruptible(&khugepaged_wait); return count; } static struct kobj_attribute scan_sleep_millisecs_attr = __ATTR(scan_sleep_millisecs, 0644, scan_sleep_millisecs_show, scan_sleep_millisecs_store); static ssize_t alloc_sleep_millisecs_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return sprintf(buf, "%u ", khugepaged_alloc_sleep_millisecs); } static ssize_t alloc_sleep_millisecs_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t count) { unsigned long msecs; int err; err = kstrtoul(buf, 10, &msecs); if (err || msecs > UINT_MAX) return -EINVAL; khugepaged_alloc_sleep_millisecs = msecs; khugepaged_sleep_expire = 0; wake_up_interruptible(&khugepaged_wait); return count; } static struct kobj_attribute alloc_sleep_millisecs_attr = __ATTR(alloc_sleep_millisecs, 0644, alloc_sleep_millisecs_show, alloc_sleep_millisecs_store); static ssize_t pages_to_scan_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return sprintf(buf, "%u ", khugepaged_pages_to_scan); } static ssize_t pages_to_scan_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t count) { int err; unsigned long pages; err = kstrtoul(buf, 10, &pages); if (err || !pages || pages > UINT_MAX) return -EINVAL; khugepaged_pages_to_scan = pages; return count; } static struct kobj_attribute pages_to_scan_attr = __ATTR(pages_to_scan, 0644, pages_to_scan_show, pages_to_scan_store); static ssize_t pages_collapsed_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return sprintf(buf, "%u ", khugepaged_pages_collapsed); } static struct kobj_attribute pages_collapsed_attr = __ATTR_RO(pages_collapsed); static ssize_t full_scans_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return sprintf(buf, "%u ", khugepaged_full_scans); } static struct kobj_attribute full_scans_attr = __ATTR_RO(full_scans); static ssize_t khugepaged_defrag_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return single_hugepage_flag_show(kobj, attr, buf, TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG); } static ssize_t khugepaged_defrag_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t count) { return single_hugepage_flag_store(kobj, attr, buf, count, TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG); } static struct kobj_attribute khugepaged_defrag_attr = __ATTR(defrag, 0644, khugepaged_defrag_show, khugepaged_defrag_store); /* * max_ptes_none controls if khugepaged should collapse hugepages over * any unmapped ptes in turn potentially increasing the memory * footprint of the vmas. When max_ptes_none is 0 khugepaged will not * reduce the available free memory in the system as it * runs. Increasing max_ptes_none will instead potentially reduce the * free memory in the system during the khugepaged scan. */ static ssize_t khugepaged_max_ptes_none_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return sprintf(buf, "%u ", khugepaged_max_ptes_none); } static ssize_t khugepaged_max_ptes_none_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t count) { int err; unsigned long max_ptes_none; err = kstrtoul(buf, 10, &max_ptes_none); if (err || max_ptes_none > HPAGE_PMD_NR-1) return -EINVAL; khugepaged_max_ptes_none = max_ptes_none; return count; } static struct kobj_attribute khugepaged_max_ptes_none_attr = __ATTR(max_ptes_none, 0644, khugepaged_max_ptes_none_show, khugepaged_max_ptes_none_store); static ssize_t khugepaged_max_ptes_swap_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { return sprintf(buf, "%u ", khugepaged_max_ptes_swap); } static ssize_t khugepaged_max_ptes_swap_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t count) { int err; unsigned long max_ptes_swap; err = kstrtoul(buf, 10, &max_ptes_swap); if (err || max_ptes_swap > HPAGE_PMD_NR-1) return -EINVAL; khugepaged_max_ptes_swap = max_ptes_swap; return count; } static struct kobj_attribute khugepaged_max_ptes_swap_attr = __ATTR(max_ptes_swap, 0644, khugepaged_max_ptes_swap_show, khugepaged_max_ptes_swap_store); static struct attribute *khugepaged_attr[] = { &khugepaged_defrag_attr.attr, &khugepaged_max_ptes_none_attr.attr, &pages_to_scan_attr.attr, &pages_collapsed_attr.attr, &full_scans_attr.attr, &scan_sleep_millisecs_attr.attr, &alloc_sleep_millisecs_attr.attr, &khugepaged_max_ptes_swap_attr.attr, NULL, }; struct attribute_group khugepaged_attr_group = { .attrs = khugepaged_attr, .name = "khugepaged", }; |
e1465d125 mm, thp: propagat... |
313 |
#endif /* CONFIG_SYSFS */ |
b46e756f5 thp: extract khug... |
314 |
|
f3f0e1d21 khugepaged: add s... |
315 |
#define VM_NO_KHUGEPAGED (VM_SPECIAL | VM_HUGETLB) |
b46e756f5 thp: extract khug... |
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 |
int hugepage_madvise(struct vm_area_struct *vma, unsigned long *vm_flags, int advice) { switch (advice) { case MADV_HUGEPAGE: #ifdef CONFIG_S390 /* * qemu blindly sets MADV_HUGEPAGE on all allocations, but s390 * can't handle this properly after s390_enable_sie, so we simply * ignore the madvise to prevent qemu from causing a SIGSEGV. */ if (mm_has_pgste(vma->vm_mm)) return 0; #endif *vm_flags &= ~VM_NOHUGEPAGE; *vm_flags |= VM_HUGEPAGE; /* * If the vma become good for khugepaged to scan, * register it here without waiting a page fault that * may not happen any time soon. */ if (!(*vm_flags & VM_NO_KHUGEPAGED) && khugepaged_enter_vma_merge(vma, *vm_flags)) return -ENOMEM; break; case MADV_NOHUGEPAGE: *vm_flags &= ~VM_HUGEPAGE; *vm_flags |= VM_NOHUGEPAGE; /* * Setting VM_NOHUGEPAGE will prevent khugepaged from scanning * this vma even if we leave the mm registered in khugepaged if * it got registered before VM_NOHUGEPAGE was set. */ break; } return 0; } int __init khugepaged_init(void) { mm_slot_cache = kmem_cache_create("khugepaged_mm_slot", sizeof(struct mm_slot), __alignof__(struct mm_slot), 0, NULL); if (!mm_slot_cache) return -ENOMEM; khugepaged_pages_to_scan = HPAGE_PMD_NR * 8; khugepaged_max_ptes_none = HPAGE_PMD_NR - 1; khugepaged_max_ptes_swap = HPAGE_PMD_NR / 8; return 0; } void __init khugepaged_destroy(void) { kmem_cache_destroy(mm_slot_cache); } static inline struct mm_slot *alloc_mm_slot(void) { if (!mm_slot_cache) /* initialization failed */ return NULL; return kmem_cache_zalloc(mm_slot_cache, GFP_KERNEL); } static inline void free_mm_slot(struct mm_slot *mm_slot) { kmem_cache_free(mm_slot_cache, mm_slot); } static struct mm_slot *get_mm_slot(struct mm_struct *mm) { struct mm_slot *mm_slot; hash_for_each_possible(mm_slots_hash, mm_slot, hash, (unsigned long)mm) if (mm == mm_slot->mm) return mm_slot; return NULL; } static void insert_to_mm_slots_hash(struct mm_struct *mm, struct mm_slot *mm_slot) { mm_slot->mm = mm; hash_add(mm_slots_hash, &mm_slot->hash, (long)mm); } static inline int khugepaged_test_exit(struct mm_struct *mm) { return atomic_read(&mm->mm_users) == 0; } |
50f8b92f2 mm: thp: pass cor... |
410 411 |
static bool hugepage_vma_check(struct vm_area_struct *vma, unsigned long vm_flags) |
c2231020e mm: thp: register... |
412 |
{ |
50f8b92f2 mm: thp: pass cor... |
413 414 |
if ((!(vm_flags & VM_HUGEPAGE) && !khugepaged_always()) || (vm_flags & VM_NOHUGEPAGE) || |
c2231020e mm: thp: register... |
415 416 |
test_bit(MMF_DISABLE_THP, &vma->vm_mm->flags)) return false; |
99cb0dbd4 mm,thp: add read-... |
417 418 419 420 421 |
if (shmem_file(vma->vm_file) || (IS_ENABLED(CONFIG_READ_ONLY_THP_FOR_FS) && vma->vm_file && (vm_flags & VM_DENYWRITE))) { |
c2231020e mm: thp: register... |
422 423 424 425 426 427 428 429 430 |
if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGE_PAGECACHE)) return false; return IS_ALIGNED((vma->vm_start >> PAGE_SHIFT) - vma->vm_pgoff, HPAGE_PMD_NR); } if (!vma->anon_vma || vma->vm_ops) return false; if (is_vma_temporary_stack(vma)) return false; |
50f8b92f2 mm: thp: pass cor... |
431 |
return !(vm_flags & VM_NO_KHUGEPAGED); |
c2231020e mm: thp: register... |
432 |
} |
b46e756f5 thp: extract khug... |
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 |
int __khugepaged_enter(struct mm_struct *mm) { struct mm_slot *mm_slot; int wakeup; mm_slot = alloc_mm_slot(); if (!mm_slot) return -ENOMEM; /* __khugepaged_exit() must not run from under us */ VM_BUG_ON_MM(khugepaged_test_exit(mm), mm); if (unlikely(test_and_set_bit(MMF_VM_HUGEPAGE, &mm->flags))) { free_mm_slot(mm_slot); return 0; } spin_lock(&khugepaged_mm_lock); insert_to_mm_slots_hash(mm, mm_slot); /* * Insert just behind the scanning cursor, to let the area settle * down a little. */ wakeup = list_empty(&khugepaged_scan.mm_head); list_add_tail(&mm_slot->mm_node, &khugepaged_scan.mm_head); spin_unlock(&khugepaged_mm_lock); |
f1f100764 mm: add new mmgra... |
458 |
mmgrab(mm); |
b46e756f5 thp: extract khug... |
459 460 461 462 463 464 465 466 467 468 |
if (wakeup) wake_up_interruptible(&khugepaged_wait); return 0; } int khugepaged_enter_vma_merge(struct vm_area_struct *vma, unsigned long vm_flags) { unsigned long hstart, hend; |
c2231020e mm: thp: register... |
469 470 |
/* |
99cb0dbd4 mm,thp: add read-... |
471 472 473 |
* khugepaged only supports read-only files for non-shmem files. * khugepaged does not yet work on special mappings. And * file-private shmem THP is not supported. |
c2231020e mm: thp: register... |
474 |
*/ |
50f8b92f2 mm: thp: pass cor... |
475 |
if (!hugepage_vma_check(vma, vm_flags)) |
b46e756f5 thp: extract khug... |
476 |
return 0; |
c2231020e mm: thp: register... |
477 |
|
b46e756f5 thp: extract khug... |
478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 |
hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; hend = vma->vm_end & HPAGE_PMD_MASK; if (hstart < hend) return khugepaged_enter(vma, vm_flags); return 0; } void __khugepaged_exit(struct mm_struct *mm) { struct mm_slot *mm_slot; int free = 0; spin_lock(&khugepaged_mm_lock); mm_slot = get_mm_slot(mm); if (mm_slot && khugepaged_scan.mm_slot != mm_slot) { hash_del(&mm_slot->hash); list_del(&mm_slot->mm_node); free = 1; } spin_unlock(&khugepaged_mm_lock); if (free) { clear_bit(MMF_VM_HUGEPAGE, &mm->flags); free_mm_slot(mm_slot); mmdrop(mm); } else if (mm_slot) { /* * This is required to serialize against * khugepaged_test_exit() (which is guaranteed to run * under mmap sem read mode). Stop here (after we * return all pagetables will be destroyed) until * khugepaged has finished working on the pagetables * under the mmap_sem. */ down_write(&mm->mmap_sem); up_write(&mm->mmap_sem); } } static void release_pte_page(struct page *page) { |
d44d363f6 mm: don't assume ... |
519 |
dec_node_page_state(page, NR_ISOLATED_ANON + page_is_file_cache(page)); |
b46e756f5 thp: extract khug... |
520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 |
unlock_page(page); putback_lru_page(page); } static void release_pte_pages(pte_t *pte, pte_t *_pte) { while (--_pte >= pte) { pte_t pteval = *_pte; if (!pte_none(pteval) && !is_zero_pfn(pte_pfn(pteval))) release_pte_page(pte_page(pteval)); } } static int __collapse_huge_page_isolate(struct vm_area_struct *vma, unsigned long address, pte_t *pte) { struct page *page = NULL; pte_t *_pte; |
0db501f7a mm, thp: convert ... |
539 540 |
int none_or_zero = 0, result = 0, referenced = 0; bool writable = false; |
b46e756f5 thp: extract khug... |
541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 |
for (_pte = pte; _pte < pte+HPAGE_PMD_NR; _pte++, address += PAGE_SIZE) { pte_t pteval = *_pte; if (pte_none(pteval) || (pte_present(pteval) && is_zero_pfn(pte_pfn(pteval)))) { if (!userfaultfd_armed(vma) && ++none_or_zero <= khugepaged_max_ptes_none) { continue; } else { result = SCAN_EXCEED_NONE_PTE; goto out; } } if (!pte_present(pteval)) { result = SCAN_PTE_NON_PRESENT; goto out; } page = vm_normal_page(vma, address, pteval); if (unlikely(!page)) { result = SCAN_PAGE_NULL; goto out; } |
fece2029a mm/khugepaged.c: ... |
564 565 566 567 568 |
/* TODO: teach khugepaged to collapse THP mapped with pte */ if (PageCompound(page)) { result = SCAN_PAGE_COMPOUND; goto out; } |
b46e756f5 thp: extract khug... |
569 |
VM_BUG_ON_PAGE(!PageAnon(page), page); |
b46e756f5 thp: extract khug... |
570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 |
/* * We can do it before isolate_lru_page because the * page can't be freed from under us. NOTE: PG_lock * is needed to serialize against split_huge_page * when invoked from the VM. */ if (!trylock_page(page)) { result = SCAN_PAGE_LOCK; goto out; } /* * cannot use mapcount: can't collapse if there's a gup pin. * The page must only be referenced by the scanned process * and page swap cache. */ |
2948be5ac mm: do not use do... |
587 |
if (page_count(page) != 1 + PageSwapCache(page)) { |
b46e756f5 thp: extract khug... |
588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 |
unlock_page(page); result = SCAN_PAGE_COUNT; goto out; } if (pte_write(pteval)) { writable = true; } else { if (PageSwapCache(page) && !reuse_swap_page(page, NULL)) { unlock_page(page); result = SCAN_SWAP_CACHE_PAGE; goto out; } /* * Page is not in the swap cache. It can be collapsed * into a THP. */ } /* * Isolate the page to avoid collapsing an hugepage * currently in use by the VM. */ if (isolate_lru_page(page)) { unlock_page(page); result = SCAN_DEL_PAGE_LRU; goto out; } |
d44d363f6 mm: don't assume ... |
616 617 |
inc_node_page_state(page, NR_ISOLATED_ANON + page_is_file_cache(page)); |
b46e756f5 thp: extract khug... |
618 619 |
VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageLRU(page), page); |
0db501f7a mm, thp: convert ... |
620 |
/* There should be enough young pte to collapse the page */ |
b46e756f5 thp: extract khug... |
621 622 623 |
if (pte_young(pteval) || page_is_young(page) || PageReferenced(page) || mmu_notifier_test_young(vma->vm_mm, address)) |
0db501f7a mm, thp: convert ... |
624 |
referenced++; |
b46e756f5 thp: extract khug... |
625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 |
} if (likely(writable)) { if (likely(referenced)) { result = SCAN_SUCCEED; trace_mm_collapse_huge_page_isolate(page, none_or_zero, referenced, writable, result); return 1; } } else { result = SCAN_PAGE_RO; } out: release_pte_pages(pte, _pte); trace_mm_collapse_huge_page_isolate(page, none_or_zero, referenced, writable, result); return 0; } static void __collapse_huge_page_copy(pte_t *pte, struct page *page, struct vm_area_struct *vma, unsigned long address, spinlock_t *ptl) { pte_t *_pte; |
338a16ba1 mm, thp: copying ... |
650 651 |
for (_pte = pte; _pte < pte + HPAGE_PMD_NR; _pte++, page++, address += PAGE_SIZE) { |
b46e756f5 thp: extract khug... |
652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 |
pte_t pteval = *_pte; struct page *src_page; if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) { clear_user_highpage(page, address); add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1); if (is_zero_pfn(pte_pfn(pteval))) { /* * ptl mostly unnecessary. */ spin_lock(ptl); /* * paravirt calls inside pte_clear here are * superfluous. */ pte_clear(vma->vm_mm, address, _pte); spin_unlock(ptl); } } else { src_page = pte_page(pteval); copy_user_highpage(page, src_page, address, vma); VM_BUG_ON_PAGE(page_mapcount(src_page) != 1, src_page); release_pte_page(src_page); /* * ptl mostly unnecessary, but preempt has to * be disabled to update the per-cpu stats * inside page_remove_rmap(). */ spin_lock(ptl); /* * paravirt calls inside pte_clear here are * superfluous. */ pte_clear(vma->vm_mm, address, _pte); page_remove_rmap(src_page, false); spin_unlock(ptl); free_page_and_swap_cache(src_page); } |
b46e756f5 thp: extract khug... |
690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 |
} } static void khugepaged_alloc_sleep(void) { DEFINE_WAIT(wait); add_wait_queue(&khugepaged_wait, &wait); freezable_schedule_timeout_interruptible( msecs_to_jiffies(khugepaged_alloc_sleep_millisecs)); remove_wait_queue(&khugepaged_wait, &wait); } static int khugepaged_node_load[MAX_NUMNODES]; static bool khugepaged_scan_abort(int nid) { int i; /* |
a5f5f91da mm: convert zone_... |
710 |
* If node_reclaim_mode is disabled, then no extra effort is made to |
b46e756f5 thp: extract khug... |
711 712 |
* allocate memory locally. */ |
a5f5f91da mm: convert zone_... |
713 |
if (!node_reclaim_mode) |
b46e756f5 thp: extract khug... |
714 715 716 717 718 719 720 721 722 |
return false; /* If there is a count for this node already, it must be acceptable */ if (khugepaged_node_load[nid]) return false; for (i = 0; i < MAX_NUMNODES; i++) { if (!khugepaged_node_load[i]) continue; |
a55c7454a sched/topology: I... |
723 |
if (node_distance(nid, i) > node_reclaim_distance) |
b46e756f5 thp: extract khug... |
724 725 726 727 728 729 730 731 |
return true; } return false; } /* Defrag for khugepaged will enter direct reclaim/compaction if necessary */ static inline gfp_t alloc_hugepage_khugepaged_gfpmask(void) { |
251603549 mm, thp: remove _... |
732 |
return khugepaged_defrag() ? GFP_TRANSHUGE : GFP_TRANSHUGE_LIGHT; |
b46e756f5 thp: extract khug... |
733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 |
} #ifdef CONFIG_NUMA static int khugepaged_find_target_node(void) { static int last_khugepaged_target_node = NUMA_NO_NODE; int nid, target_node = 0, max_value = 0; /* find first node with max normal pages hit */ for (nid = 0; nid < MAX_NUMNODES; nid++) if (khugepaged_node_load[nid] > max_value) { max_value = khugepaged_node_load[nid]; target_node = nid; } /* do some balance if several nodes have the same hit record */ if (target_node <= last_khugepaged_target_node) for (nid = last_khugepaged_target_node + 1; nid < MAX_NUMNODES; nid++) if (max_value == khugepaged_node_load[nid]) { target_node = nid; break; } last_khugepaged_target_node = target_node; return target_node; } static bool khugepaged_prealloc_page(struct page **hpage, bool *wait) { if (IS_ERR(*hpage)) { if (!*wait) return false; *wait = false; *hpage = NULL; khugepaged_alloc_sleep(); } else if (*hpage) { put_page(*hpage); *hpage = NULL; } return true; } static struct page * |
988ddb710 khugepaged: move ... |
779 |
khugepaged_alloc_page(struct page **hpage, gfp_t gfp, int node) |
b46e756f5 thp: extract khug... |
780 781 |
{ VM_BUG_ON_PAGE(*hpage, *hpage); |
b46e756f5 thp: extract khug... |
782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 |
*hpage = __alloc_pages_node(node, gfp, HPAGE_PMD_ORDER); if (unlikely(!*hpage)) { count_vm_event(THP_COLLAPSE_ALLOC_FAILED); *hpage = ERR_PTR(-ENOMEM); return NULL; } prep_transhuge_page(*hpage); count_vm_event(THP_COLLAPSE_ALLOC); return *hpage; } #else static int khugepaged_find_target_node(void) { return 0; } static inline struct page *alloc_khugepaged_hugepage(void) { struct page *page; page = alloc_pages(alloc_hugepage_khugepaged_gfpmask(), HPAGE_PMD_ORDER); if (page) prep_transhuge_page(page); return page; } static struct page *khugepaged_alloc_hugepage(bool *wait) { struct page *hpage; do { hpage = alloc_khugepaged_hugepage(); if (!hpage) { count_vm_event(THP_COLLAPSE_ALLOC_FAILED); if (!*wait) return NULL; *wait = false; khugepaged_alloc_sleep(); } else count_vm_event(THP_COLLAPSE_ALLOC); } while (unlikely(!hpage) && likely(khugepaged_enabled())); return hpage; } static bool khugepaged_prealloc_page(struct page **hpage, bool *wait) { if (!*hpage) *hpage = khugepaged_alloc_hugepage(wait); if (unlikely(!*hpage)) return false; return true; } static struct page * |
988ddb710 khugepaged: move ... |
842 |
khugepaged_alloc_page(struct page **hpage, gfp_t gfp, int node) |
b46e756f5 thp: extract khug... |
843 |
{ |
b46e756f5 thp: extract khug... |
844 845 846 847 848 |
VM_BUG_ON(!*hpage); return *hpage; } #endif |
b46e756f5 thp: extract khug... |
849 850 851 852 853 854 |
/* * If mmap_sem temporarily dropped, revalidate vma * before taking mmap_sem. * Return 0 if succeeds, otherwise return none-zero * value (scan code). */ |
c131f751a khugepaged: fix u... |
855 856 |
static int hugepage_vma_revalidate(struct mm_struct *mm, unsigned long address, struct vm_area_struct **vmap) |
b46e756f5 thp: extract khug... |
857 858 859 860 861 862 |
{ struct vm_area_struct *vma; unsigned long hstart, hend; if (unlikely(khugepaged_test_exit(mm))) return SCAN_ANY_PROCESS; |
c131f751a khugepaged: fix u... |
863 |
*vmap = vma = find_vma(mm, address); |
b46e756f5 thp: extract khug... |
864 865 866 867 868 869 870 |
if (!vma) return SCAN_VMA_NULL; hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; hend = vma->vm_end & HPAGE_PMD_MASK; if (address < hstart || address + HPAGE_PMD_SIZE > hend) return SCAN_ADDRESS_RANGE; |
50f8b92f2 mm: thp: pass cor... |
871 |
if (!hugepage_vma_check(vma, vma->vm_flags)) |
b46e756f5 thp: extract khug... |
872 |
return SCAN_VMA_CHECK; |
40c5836b4 khugepaged: fix n... |
873 874 875 |
/* Anon VMA expected */ if (!vma->anon_vma || vma->vm_ops) return SCAN_VMA_CHECK; |
b46e756f5 thp: extract khug... |
876 877 878 879 880 881 882 883 884 885 886 887 888 |
return 0; } /* * Bring missing pages in from swap, to complete THP collapse. * Only done if khugepaged_scan_pmd believes it is worthwhile. * * Called and returns without pte mapped or spinlocks held, * but with mmap_sem held to protect against vma changes. */ static bool __collapse_huge_page_swapin(struct mm_struct *mm, struct vm_area_struct *vma, |
0db501f7a mm, thp: convert ... |
889 890 |
unsigned long address, pmd_t *pmd, int referenced) |
b46e756f5 thp: extract khug... |
891 |
{ |
2b7403035 mm: Change return... |
892 893 |
int swapped_in = 0; vm_fault_t ret = 0; |
82b0f8c39 mm: join struct f... |
894 |
struct vm_fault vmf = { |
b46e756f5 thp: extract khug... |
895 896 897 898 |
.vma = vma, .address = address, .flags = FAULT_FLAG_ALLOW_RETRY, .pmd = pmd, |
0721ec8bc mm: use pgoff in ... |
899 |
.pgoff = linear_page_index(vma, address), |
b46e756f5 thp: extract khug... |
900 |
}; |
982785c6b mm, thp: fix leak... |
901 902 903 904 905 |
/* we only decide to swapin, if there is enough young ptes */ if (referenced < HPAGE_PMD_NR/2) { trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0); return false; } |
82b0f8c39 mm: join struct f... |
906 907 908 |
vmf.pte = pte_offset_map(pmd, address); for (; vmf.address < address + HPAGE_PMD_NR*PAGE_SIZE; vmf.pte++, vmf.address += PAGE_SIZE) { |
2994302bc mm: add orig_pte ... |
909 910 |
vmf.orig_pte = *vmf.pte; if (!is_swap_pte(vmf.orig_pte)) |
b46e756f5 thp: extract khug... |
911 912 |
continue; swapped_in++; |
2994302bc mm: add orig_pte ... |
913 |
ret = do_swap_page(&vmf); |
0db501f7a mm, thp: convert ... |
914 |
|
b46e756f5 thp: extract khug... |
915 916 917 |
/* do_swap_page returns VM_FAULT_RETRY with released mmap_sem */ if (ret & VM_FAULT_RETRY) { down_read(&mm->mmap_sem); |
82b0f8c39 mm: join struct f... |
918 |
if (hugepage_vma_revalidate(mm, address, &vmf.vma)) { |
47f863ea2 mm, thp: fix comm... |
919 |
/* vma is no longer available, don't continue to swapin */ |
0db501f7a mm, thp: convert ... |
920 |
trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0); |
b46e756f5 thp: extract khug... |
921 |
return false; |
47f863ea2 mm, thp: fix comm... |
922 |
} |
b46e756f5 thp: extract khug... |
923 |
/* check if the pmd is still valid */ |
835152a25 mm/khugepaged: ad... |
924 925 |
if (mm_find_pmd(mm, address) != pmd) { trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0); |
b46e756f5 thp: extract khug... |
926 |
return false; |
835152a25 mm/khugepaged: ad... |
927 |
} |
b46e756f5 thp: extract khug... |
928 929 |
} if (ret & VM_FAULT_ERROR) { |
0db501f7a mm, thp: convert ... |
930 |
trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0); |
b46e756f5 thp: extract khug... |
931 932 933 |
return false; } /* pte is unmapped now, we need to map it */ |
82b0f8c39 mm: join struct f... |
934 |
vmf.pte = pte_offset_map(pmd, vmf.address); |
b46e756f5 thp: extract khug... |
935 |
} |
82b0f8c39 mm: join struct f... |
936 937 |
vmf.pte--; pte_unmap(vmf.pte); |
0db501f7a mm, thp: convert ... |
938 |
trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 1); |
b46e756f5 thp: extract khug... |
939 940 941 942 943 944 |
return true; } static void collapse_huge_page(struct mm_struct *mm, unsigned long address, struct page **hpage, |
0db501f7a mm, thp: convert ... |
945 |
int node, int referenced) |
b46e756f5 thp: extract khug... |
946 947 948 949 950 951 952 953 |
{ pmd_t *pmd, _pmd; pte_t *pte; pgtable_t pgtable; struct page *new_page; spinlock_t *pmd_ptl, *pte_ptl; int isolated = 0, result = 0; struct mem_cgroup *memcg; |
c131f751a khugepaged: fix u... |
954 |
struct vm_area_struct *vma; |
ac46d4f3c mm/mmu_notifier: ... |
955 |
struct mmu_notifier_range range; |
b46e756f5 thp: extract khug... |
956 957 958 959 960 |
gfp_t gfp; VM_BUG_ON(address & ~HPAGE_PMD_MASK); /* Only allocate from the target node */ |
41b6167e8 mm: get rid of __... |
961 |
gfp = alloc_hugepage_khugepaged_gfpmask() | __GFP_THISNODE; |
b46e756f5 thp: extract khug... |
962 |
|
988ddb710 khugepaged: move ... |
963 964 965 966 967 968 969 970 |
/* * Before allocating the hugepage, release the mmap_sem read lock. * The allocation can take potentially a long time if it involves * sync compaction, and we do not need to hold the mmap_sem during * that. We will recheck the vma after taking it again in write mode. */ up_read(&mm->mmap_sem); new_page = khugepaged_alloc_page(hpage, gfp, node); |
b46e756f5 thp: extract khug... |
971 972 973 974 |
if (!new_page) { result = SCAN_ALLOC_HUGE_PAGE_FAIL; goto out_nolock; } |
2a70f6a76 memcg, thp: do no... |
975 |
if (unlikely(mem_cgroup_try_charge(new_page, mm, gfp, &memcg, true))) { |
b46e756f5 thp: extract khug... |
976 977 978 979 980 |
result = SCAN_CGROUP_CHARGE_FAIL; goto out_nolock; } down_read(&mm->mmap_sem); |
c131f751a khugepaged: fix u... |
981 |
result = hugepage_vma_revalidate(mm, address, &vma); |
b46e756f5 thp: extract khug... |
982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 |
if (result) { mem_cgroup_cancel_charge(new_page, memcg, true); up_read(&mm->mmap_sem); goto out_nolock; } pmd = mm_find_pmd(mm, address); if (!pmd) { result = SCAN_PMD_NULL; mem_cgroup_cancel_charge(new_page, memcg, true); up_read(&mm->mmap_sem); goto out_nolock; } /* * __collapse_huge_page_swapin always returns with mmap_sem locked. |
47f863ea2 mm, thp: fix comm... |
998 |
* If it fails, we release mmap_sem and jump out_nolock. |
b46e756f5 thp: extract khug... |
999 1000 |
* Continuing to collapse causes inconsistency. */ |
0db501f7a mm, thp: convert ... |
1001 |
if (!__collapse_huge_page_swapin(mm, vma, address, pmd, referenced)) { |
b46e756f5 thp: extract khug... |
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 |
mem_cgroup_cancel_charge(new_page, memcg, true); up_read(&mm->mmap_sem); goto out_nolock; } up_read(&mm->mmap_sem); /* * Prevent all access to pagetables with the exception of * gup_fast later handled by the ptep_clear_flush and the VM * handled by the anon_vma lock + PG_lock. */ down_write(&mm->mmap_sem); |
59ea6d06c coredump: fix rac... |
1014 1015 1016 |
result = SCAN_ANY_PROCESS; if (!mmget_still_valid(mm)) goto out; |
c131f751a khugepaged: fix u... |
1017 |
result = hugepage_vma_revalidate(mm, address, &vma); |
b46e756f5 thp: extract khug... |
1018 1019 1020 1021 1022 1023 1024 |
if (result) goto out; /* check if the pmd is still valid */ if (mm_find_pmd(mm, address) != pmd) goto out; anon_vma_lock_write(vma->anon_vma); |
7269f9999 mm/mmu_notifier: ... |
1025 |
mmu_notifier_range_init(&range, MMU_NOTIFY_CLEAR, 0, NULL, mm, |
6f4f13e8d mm/mmu_notifier: ... |
1026 |
address, address + HPAGE_PMD_SIZE); |
ac46d4f3c mm/mmu_notifier: ... |
1027 |
mmu_notifier_invalidate_range_start(&range); |
ec649c9d4 mm/khugepaged: fi... |
1028 1029 1030 |
pte = pte_offset_map(pmd, address); pte_ptl = pte_lockptr(mm, pmd); |
b46e756f5 thp: extract khug... |
1031 1032 1033 1034 1035 1036 1037 1038 1039 |
pmd_ptl = pmd_lock(mm, pmd); /* probably unnecessary */ /* * After this gup_fast can't run anymore. This also removes * any huge TLB entry from the CPU so we won't allow * huge and small TLB entries for the same virtual address * to avoid the risk of CPU bugs in that area. */ _pmd = pmdp_collapse_flush(vma, address, pmd); spin_unlock(pmd_ptl); |
ac46d4f3c mm/mmu_notifier: ... |
1040 |
mmu_notifier_invalidate_range_end(&range); |
b46e756f5 thp: extract khug... |
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 1071 1072 1073 |
spin_lock(pte_ptl); isolated = __collapse_huge_page_isolate(vma, address, pte); spin_unlock(pte_ptl); if (unlikely(!isolated)) { pte_unmap(pte); spin_lock(pmd_ptl); BUG_ON(!pmd_none(*pmd)); /* * We can only use set_pmd_at when establishing * hugepmds and never for establishing regular pmds that * points to regular pagetables. Use pmd_populate for that */ pmd_populate(mm, pmd, pmd_pgtable(_pmd)); spin_unlock(pmd_ptl); anon_vma_unlock_write(vma->anon_vma); result = SCAN_FAIL; goto out; } /* * All pages are isolated and locked so anon_vma rmap * can't run anymore. */ anon_vma_unlock_write(vma->anon_vma); __collapse_huge_page_copy(pte, new_page, vma, address, pte_ptl); pte_unmap(pte); __SetPageUptodate(new_page); pgtable = pmd_pgtable(_pmd); _pmd = mk_huge_pmd(new_page, vma->vm_page_prot); |
f55e1014f Revert "mm, thp: ... |
1074 |
_pmd = maybe_pmd_mkwrite(pmd_mkdirty(_pmd), vma); |
b46e756f5 thp: extract khug... |
1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 |
/* * spin_lock() below is not the equivalent of smp_wmb(), so * this is needed to avoid the copy_huge_page writes to become * visible after the set_pmd_at() write. */ smp_wmb(); spin_lock(pmd_ptl); BUG_ON(!pmd_none(*pmd)); page_add_new_anon_rmap(new_page, vma, address, true); mem_cgroup_commit_charge(new_page, memcg, false, true); |
1ff9e6e17 mm: memcontrol: e... |
1087 |
count_memcg_events(memcg, THP_COLLAPSE_ALLOC, 1); |
b46e756f5 thp: extract khug... |
1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 |
lru_cache_add_active_or_unevictable(new_page, vma); pgtable_trans_huge_deposit(mm, pmd, pgtable); set_pmd_at(mm, address, pmd, _pmd); update_mmu_cache_pmd(vma, address, pmd); spin_unlock(pmd_ptl); *hpage = NULL; khugepaged_pages_collapsed++; result = SCAN_SUCCEED; out_up_write: up_write(&mm->mmap_sem); out_nolock: trace_mm_collapse_huge_page(mm, isolated, result); return; out: mem_cgroup_cancel_charge(new_page, memcg, true); goto out_up_write; } static int khugepaged_scan_pmd(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, struct page **hpage) { pmd_t *pmd; pte_t *pte, *_pte; |
0db501f7a mm, thp: convert ... |
1115 |
int ret = 0, none_or_zero = 0, result = 0, referenced = 0; |
b46e756f5 thp: extract khug... |
1116 1117 1118 1119 |
struct page *page = NULL; unsigned long _address; spinlock_t *ptl; int node = NUMA_NO_NODE, unmapped = 0; |
0db501f7a mm, thp: convert ... |
1120 |
bool writable = false; |
b46e756f5 thp: extract khug... |
1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 |
VM_BUG_ON(address & ~HPAGE_PMD_MASK); pmd = mm_find_pmd(mm, address); if (!pmd) { result = SCAN_PMD_NULL; goto out; } memset(khugepaged_node_load, 0, sizeof(khugepaged_node_load)); pte = pte_offset_map_lock(mm, pmd, address, &ptl); for (_address = address, _pte = pte; _pte < pte+HPAGE_PMD_NR; _pte++, _address += PAGE_SIZE) { pte_t pteval = *_pte; if (is_swap_pte(pteval)) { if (++unmapped <= khugepaged_max_ptes_swap) { continue; } else { result = SCAN_EXCEED_SWAP_PTE; goto out_unmap; } } if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) { if (!userfaultfd_armed(vma) && ++none_or_zero <= khugepaged_max_ptes_none) { continue; } else { result = SCAN_EXCEED_NONE_PTE; goto out_unmap; } } if (!pte_present(pteval)) { result = SCAN_PTE_NON_PRESENT; goto out_unmap; } if (pte_write(pteval)) writable = true; page = vm_normal_page(vma, _address, pteval); if (unlikely(!page)) { result = SCAN_PAGE_NULL; goto out_unmap; } /* TODO: teach khugepaged to collapse THP mapped with pte */ if (PageCompound(page)) { result = SCAN_PAGE_COMPOUND; goto out_unmap; } /* * Record which node the original page is from and save this * information to khugepaged_node_load[]. * Khupaged will allocate hugepage from the node has the max * hit record. */ node = page_to_nid(page); if (khugepaged_scan_abort(node)) { result = SCAN_SCAN_ABORT; goto out_unmap; } khugepaged_node_load[node]++; if (!PageLRU(page)) { result = SCAN_PAGE_LRU; goto out_unmap; } if (PageLocked(page)) { result = SCAN_PAGE_LOCK; goto out_unmap; } if (!PageAnon(page)) { result = SCAN_PAGE_ANON; goto out_unmap; } /* * cannot use mapcount: can't collapse if there's a gup pin. * The page must only be referenced by the scanned process * and page swap cache. */ |
2948be5ac mm: do not use do... |
1201 |
if (page_count(page) != 1 + PageSwapCache(page)) { |
b46e756f5 thp: extract khug... |
1202 1203 1204 1205 1206 1207 |
result = SCAN_PAGE_COUNT; goto out_unmap; } if (pte_young(pteval) || page_is_young(page) || PageReferenced(page) || mmu_notifier_test_young(vma->vm_mm, address)) |
0db501f7a mm, thp: convert ... |
1208 |
referenced++; |
b46e756f5 thp: extract khug... |
1209 1210 1211 1212 1213 1214 |
} if (writable) { if (referenced) { result = SCAN_SUCCEED; ret = 1; } else { |
0db501f7a mm, thp: convert ... |
1215 |
result = SCAN_LACK_REFERENCED_PAGE; |
b46e756f5 thp: extract khug... |
1216 1217 1218 1219 1220 1221 1222 1223 1224 |
} } else { result = SCAN_PAGE_RO; } out_unmap: pte_unmap_unlock(pte, ptl); if (ret) { node = khugepaged_find_target_node(); /* collapse_huge_page will return with the mmap_sem released */ |
c131f751a khugepaged: fix u... |
1225 |
collapse_huge_page(mm, address, hpage, node, referenced); |
b46e756f5 thp: extract khug... |
1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 |
} out: trace_mm_khugepaged_scan_pmd(mm, page, writable, referenced, none_or_zero, result, unmapped); return ret; } static void collect_mm_slot(struct mm_slot *mm_slot) { struct mm_struct *mm = mm_slot->mm; |
35f3aa39f mm: Replace spin_... |
1236 |
lockdep_assert_held(&khugepaged_mm_lock); |
b46e756f5 thp: extract khug... |
1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 |
if (khugepaged_test_exit(mm)) { /* free mm_slot */ hash_del(&mm_slot->hash); list_del(&mm_slot->mm_node); /* * Not strictly needed because the mm exited already. * * clear_bit(MMF_VM_HUGEPAGE, &mm->flags); */ /* khugepaged_mm_lock actually not necessary for the below */ free_mm_slot(mm_slot); mmdrop(mm); } } |
e496cf3d7 thp: introduce CO... |
1254 |
#if defined(CONFIG_SHMEM) && defined(CONFIG_TRANSPARENT_HUGE_PAGECACHE) |
27e1f8273 khugepaged: enabl... |
1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 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 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 |
/* * Notify khugepaged that given addr of the mm is pte-mapped THP. Then * khugepaged should try to collapse the page table. */ static int khugepaged_add_pte_mapped_thp(struct mm_struct *mm, unsigned long addr) { struct mm_slot *mm_slot; VM_BUG_ON(addr & ~HPAGE_PMD_MASK); spin_lock(&khugepaged_mm_lock); mm_slot = get_mm_slot(mm); if (likely(mm_slot && mm_slot->nr_pte_mapped_thp < MAX_PTE_MAPPED_THP)) mm_slot->pte_mapped_thp[mm_slot->nr_pte_mapped_thp++] = addr; spin_unlock(&khugepaged_mm_lock); return 0; } /** * Try to collapse a pte-mapped THP for mm at address haddr. * * This function checks whether all the PTEs in the PMD are pointing to the * right THP. If so, retract the page table so the THP can refault in with * as pmd-mapped. */ void collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr) { unsigned long haddr = addr & HPAGE_PMD_MASK; struct vm_area_struct *vma = find_vma(mm, haddr); struct page *hpage = NULL; pte_t *start_pte, *pte; pmd_t *pmd, _pmd; spinlock_t *ptl; int count = 0; int i; if (!vma || !vma->vm_file || vma->vm_start > haddr || vma->vm_end < haddr + HPAGE_PMD_SIZE) return; /* * This vm_flags may not have VM_HUGEPAGE if the page was not * collapsed by this mm. But we can still collapse if the page is * the valid THP. Add extra VM_HUGEPAGE so hugepage_vma_check() * will not fail the vma for missing VM_HUGEPAGE */ if (!hugepage_vma_check(vma, vma->vm_flags | VM_HUGEPAGE)) return; pmd = mm_find_pmd(mm, haddr); if (!pmd) return; start_pte = pte_offset_map_lock(mm, pmd, haddr, &ptl); /* step 1: check all mapped PTEs are to the right huge page */ for (i = 0, addr = haddr, pte = start_pte; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE, pte++) { struct page *page; /* empty pte, skip */ if (pte_none(*pte)) continue; /* page swapped out, abort */ if (!pte_present(*pte)) goto abort; page = vm_normal_page(vma, addr, *pte); if (!page || !PageCompound(page)) goto abort; if (!hpage) { hpage = compound_head(page); /* * The mapping of the THP should not change. * * Note that uprobe, debugger, or MAP_PRIVATE may * change the page table, but the new page will * not pass PageCompound() check. */ if (WARN_ON(hpage->mapping != vma->vm_file->f_mapping)) goto abort; } /* * Confirm the page maps to the correct subpage. * * Note that uprobe, debugger, or MAP_PRIVATE may change * the page table, but the new page will not pass * PageCompound() check. */ if (WARN_ON(hpage + i != page)) goto abort; count++; } /* step 2: adjust rmap */ for (i = 0, addr = haddr, pte = start_pte; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE, pte++) { struct page *page; if (pte_none(*pte)) continue; page = vm_normal_page(vma, addr, *pte); page_remove_rmap(page, false); } pte_unmap_unlock(start_pte, ptl); /* step 3: set proper refcount and mm_counters. */ if (hpage) { page_ref_sub(hpage, count); add_mm_counter(vma->vm_mm, mm_counter_file(hpage), -count); } /* step 4: collapse pmd */ ptl = pmd_lock(vma->vm_mm, pmd); _pmd = pmdp_collapse_flush(vma, addr, pmd); spin_unlock(ptl); mm_dec_nr_ptes(mm); pte_free(mm, pmd_pgtable(_pmd)); return; abort: pte_unmap_unlock(start_pte, ptl); } static int khugepaged_collapse_pte_mapped_thps(struct mm_slot *mm_slot) { struct mm_struct *mm = mm_slot->mm; int i; if (likely(mm_slot->nr_pte_mapped_thp == 0)) return 0; if (!down_write_trylock(&mm->mmap_sem)) return -EBUSY; if (unlikely(khugepaged_test_exit(mm))) goto out; for (i = 0; i < mm_slot->nr_pte_mapped_thp; i++) collapse_pte_mapped_thp(mm, mm_slot->pte_mapped_thp[i]); out: mm_slot->nr_pte_mapped_thp = 0; up_write(&mm->mmap_sem); return 0; } |
f3f0e1d21 khugepaged: add s... |
1407 1408 1409 1410 1411 1412 1413 1414 |
static void retract_page_tables(struct address_space *mapping, pgoff_t pgoff) { struct vm_area_struct *vma; unsigned long addr; pmd_t *pmd, _pmd; i_mmap_lock_write(mapping); vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) { |
27e1f8273 khugepaged: enabl... |
1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 |
/* * Check vma->anon_vma to exclude MAP_PRIVATE mappings that * got written to. These VMAs are likely not worth investing * down_write(mmap_sem) as PMD-mapping is likely to be split * later. * * Not that vma->anon_vma check is racy: it can be set up after * the check but before we took mmap_sem by the fault path. * But page lock would prevent establishing any new ptes of the * page, so we are safe. * * An alternative would be drop the check, but check that page * table is clear before calling pmdp_collapse_flush() under * ptl. It has higher chance to recover THP for the VMA, but * has higher cost too. */ |
f3f0e1d21 khugepaged: add s... |
1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 |
if (vma->anon_vma) continue; addr = vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT); if (addr & ~HPAGE_PMD_MASK) continue; if (vma->vm_end < addr + HPAGE_PMD_SIZE) continue; pmd = mm_find_pmd(vma->vm_mm, addr); if (!pmd) continue; /* * We need exclusive mmap_sem to retract page table. |
27e1f8273 khugepaged: enabl... |
1443 1444 1445 1446 |
* * We use trylock due to lock inversion: we need to acquire * mmap_sem while holding page lock. Fault path does it in * reverse order. Trylock is a way to avoid deadlock. |
f3f0e1d21 khugepaged: add s... |
1447 1448 1449 1450 1451 1452 1453 |
*/ if (down_write_trylock(&vma->vm_mm->mmap_sem)) { spinlock_t *ptl = pmd_lock(vma->vm_mm, pmd); /* assume page table is clear */ _pmd = pmdp_collapse_flush(vma, addr, pmd); spin_unlock(ptl); up_write(&vma->vm_mm->mmap_sem); |
c4812909f mm: introduce wra... |
1454 |
mm_dec_nr_ptes(vma->vm_mm); |
d670ffd87 mm/thp/pagecache/... |
1455 |
pte_free(vma->vm_mm, pmd_pgtable(_pmd)); |
27e1f8273 khugepaged: enabl... |
1456 1457 1458 |
} else { /* Try again later */ khugepaged_add_pte_mapped_thp(vma->vm_mm, addr); |
f3f0e1d21 khugepaged: add s... |
1459 1460 1461 1462 1463 1464 |
} } i_mmap_unlock_write(mapping); } /** |
99cb0dbd4 mm,thp: add read-... |
1465 |
* collapse_file - collapse filemap/tmpfs/shmem pages into huge one. |
f3f0e1d21 khugepaged: add s... |
1466 1467 |
* * Basic scheme is simple, details are more complex: |
87c460a0b mm/khugepaged: co... |
1468 |
* - allocate and lock a new huge page; |
77da9389b mm: Convert colla... |
1469 |
* - scan page cache replacing old pages with the new one |
99cb0dbd4 mm,thp: add read-... |
1470 |
* + swap/gup in pages if necessary; |
f3f0e1d21 khugepaged: add s... |
1471 |
* + fill in gaps; |
77da9389b mm: Convert colla... |
1472 1473 |
* + keep old pages around in case rollback is required; * - if replacing succeeds: |
f3f0e1d21 khugepaged: add s... |
1474 1475 |
* + copy data over; * + free old pages; |
87c460a0b mm/khugepaged: co... |
1476 |
* + unlock huge page; |
f3f0e1d21 khugepaged: add s... |
1477 1478 |
* - if replacing failed; * + put all pages back and unfreeze them; |
77da9389b mm: Convert colla... |
1479 |
* + restore gaps in the page cache; |
87c460a0b mm/khugepaged: co... |
1480 |
* + unlock and free huge page; |
f3f0e1d21 khugepaged: add s... |
1481 |
*/ |
579c571e2 khugepaged: renam... |
1482 1483 |
static void collapse_file(struct mm_struct *mm, struct file *file, pgoff_t start, |
f3f0e1d21 khugepaged: add s... |
1484 1485 |
struct page **hpage, int node) { |
579c571e2 khugepaged: renam... |
1486 |
struct address_space *mapping = file->f_mapping; |
f3f0e1d21 khugepaged: add s... |
1487 |
gfp_t gfp; |
77da9389b mm: Convert colla... |
1488 |
struct page *new_page; |
f3f0e1d21 khugepaged: add s... |
1489 1490 1491 |
struct mem_cgroup *memcg; pgoff_t index, end = start + HPAGE_PMD_NR; LIST_HEAD(pagelist); |
77da9389b mm: Convert colla... |
1492 |
XA_STATE_ORDER(xas, &mapping->i_pages, start, HPAGE_PMD_ORDER); |
f3f0e1d21 khugepaged: add s... |
1493 |
int nr_none = 0, result = SCAN_SUCCEED; |
99cb0dbd4 mm,thp: add read-... |
1494 |
bool is_shmem = shmem_file(file); |
f3f0e1d21 khugepaged: add s... |
1495 |
|
99cb0dbd4 mm,thp: add read-... |
1496 |
VM_BUG_ON(!IS_ENABLED(CONFIG_READ_ONLY_THP_FOR_FS) && !is_shmem); |
f3f0e1d21 khugepaged: add s... |
1497 1498 1499 |
VM_BUG_ON(start & (HPAGE_PMD_NR - 1)); /* Only allocate from the target node */ |
41b6167e8 mm: get rid of __... |
1500 |
gfp = alloc_hugepage_khugepaged_gfpmask() | __GFP_THISNODE; |
f3f0e1d21 khugepaged: add s... |
1501 1502 1503 1504 1505 1506 |
new_page = khugepaged_alloc_page(hpage, gfp, node); if (!new_page) { result = SCAN_ALLOC_HUGE_PAGE_FAIL; goto out; } |
2a70f6a76 memcg, thp: do no... |
1507 |
if (unlikely(mem_cgroup_try_charge(new_page, mm, gfp, &memcg, true))) { |
f3f0e1d21 khugepaged: add s... |
1508 1509 1510 |
result = SCAN_CGROUP_CHARGE_FAIL; goto out; } |
95feeabb7 mm/khugepaged: fi... |
1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 |
/* This will be less messy when we use multi-index entries */ do { xas_lock_irq(&xas); xas_create_range(&xas); if (!xas_error(&xas)) break; xas_unlock_irq(&xas); if (!xas_nomem(&xas, GFP_KERNEL)) { mem_cgroup_cancel_charge(new_page, memcg, true); result = SCAN_FAIL; goto out; } } while (1); |
042a30824 mm/khugepaged: mi... |
1524 |
__SetPageLocked(new_page); |
99cb0dbd4 mm,thp: add read-... |
1525 1526 |
if (is_shmem) __SetPageSwapBacked(new_page); |
f3f0e1d21 khugepaged: add s... |
1527 1528 |
new_page->index = start; new_page->mapping = mapping; |
f3f0e1d21 khugepaged: add s... |
1529 |
|
f3f0e1d21 khugepaged: add s... |
1530 |
/* |
87c460a0b mm/khugepaged: co... |
1531 1532 1533 |
* At this point the new_page is locked and not up-to-date. * It's safe to insert it into the page cache, because nobody would * be able to map it or use it in another way until we unlock it. |
f3f0e1d21 khugepaged: add s... |
1534 |
*/ |
77da9389b mm: Convert colla... |
1535 1536 1537 1538 1539 |
xas_set(&xas, start); for (index = start; index < end; index++) { struct page *page = xas_next(&xas); VM_BUG_ON(index != xas.xa_index); |
99cb0dbd4 mm,thp: add read-... |
1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 |
if (is_shmem) { if (!page) { /* * Stop if extent has been truncated or * hole-punched, and is now completely * empty. */ if (index == start) { if (!xas_next_entry(&xas, end - 1)) { result = SCAN_TRUNCATED; goto xa_locked; } xas_set(&xas, index); } if (!shmem_charge(mapping->host, 1)) { result = SCAN_FAIL; |
042a30824 mm/khugepaged: mi... |
1556 |
goto xa_locked; |
701270fa1 mm/khugepaged: co... |
1557 |
} |
99cb0dbd4 mm,thp: add read-... |
1558 1559 1560 |
xas_store(&xas, new_page); nr_none++; continue; |
701270fa1 mm/khugepaged: co... |
1561 |
} |
99cb0dbd4 mm,thp: add read-... |
1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 |
if (xa_is_value(page) || !PageUptodate(page)) { xas_unlock_irq(&xas); /* swap in or instantiate fallocated page */ if (shmem_getpage(mapping->host, index, &page, SGP_NOHUGE)) { result = SCAN_FAIL; goto xa_unlocked; } } else if (trylock_page(page)) { get_page(page); xas_unlock_irq(&xas); } else { result = SCAN_PAGE_LOCK; |
042a30824 mm/khugepaged: mi... |
1576 |
goto xa_locked; |
77da9389b mm: Convert colla... |
1577 |
} |
99cb0dbd4 mm,thp: add read-... |
1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 |
} else { /* !is_shmem */ if (!page || xa_is_value(page)) { xas_unlock_irq(&xas); page_cache_sync_readahead(mapping, &file->f_ra, file, index, PAGE_SIZE); /* drain pagevecs to help isolate_lru_page() */ lru_add_drain(); page = find_lock_page(mapping, index); if (unlikely(page == NULL)) { result = SCAN_FAIL; goto xa_unlocked; } |
99cb0dbd4 mm,thp: add read-... |
1591 1592 1593 1594 1595 1596 |
} else if (trylock_page(page)) { get_page(page); xas_unlock_irq(&xas); } else { result = SCAN_PAGE_LOCK; goto xa_locked; |
f3f0e1d21 khugepaged: add s... |
1597 |
} |
f3f0e1d21 khugepaged: add s... |
1598 1599 1600 |
} /* |
b93b01631 page cache: use x... |
1601 |
* The page must be locked, so we can drop the i_pages lock |
f3f0e1d21 khugepaged: add s... |
1602 1603 1604 |
* without racing with truncate. */ VM_BUG_ON_PAGE(!PageLocked(page), page); |
4655e5e5f mm,thp: recheck e... |
1605 1606 1607 1608 1609 1610 |
/* make sure the page is up to date */ if (unlikely(!PageUptodate(page))) { result = SCAN_FAIL; goto out_unlock; } |
06a5e1268 mm/khugepaged: co... |
1611 1612 1613 1614 1615 1616 1617 1618 1619 |
/* * If file was truncated then extended, or hole-punched, before * we locked the first page, then a THP might be there already. */ if (PageTransCompound(page)) { result = SCAN_PAGE_COMPOUND; goto out_unlock; } |
f3f0e1d21 khugepaged: add s... |
1620 1621 1622 1623 1624 |
if (page_mapping(page) != mapping) { result = SCAN_TRUNCATED; goto out_unlock; } |
f3f0e1d21 khugepaged: add s... |
1625 |
|
4655e5e5f mm,thp: recheck e... |
1626 1627 1628 1629 1630 1631 1632 1633 1634 |
if (!is_shmem && PageDirty(page)) { /* * khugepaged only works on read-only fd, so this * page is dirty because it hasn't been flushed * since first write. */ result = SCAN_FAIL; goto out_unlock; } |
f3f0e1d21 khugepaged: add s... |
1635 1636 |
if (isolate_lru_page(page)) { result = SCAN_DEL_PAGE_LRU; |
042a30824 mm/khugepaged: mi... |
1637 |
goto out_unlock; |
f3f0e1d21 khugepaged: add s... |
1638 |
} |
99cb0dbd4 mm,thp: add read-... |
1639 1640 1641 |
if (page_has_private(page) && !try_to_release_page(page, GFP_KERNEL)) { result = SCAN_PAGE_HAS_PRIVATE; |
a7ba9f247 mm,thp: stop leak... |
1642 |
putback_lru_page(page); |
99cb0dbd4 mm,thp: add read-... |
1643 1644 |
goto out_unlock; } |
f3f0e1d21 khugepaged: add s... |
1645 |
if (page_mapped(page)) |
977fbdcd5 mm: add unmap_map... |
1646 |
unmap_mapping_pages(mapping, index, 1, false); |
f3f0e1d21 khugepaged: add s... |
1647 |
|
77da9389b mm: Convert colla... |
1648 1649 |
xas_lock_irq(&xas); xas_set(&xas, index); |
f3f0e1d21 khugepaged: add s... |
1650 |
|
77da9389b mm: Convert colla... |
1651 |
VM_BUG_ON_PAGE(page != xas_load(&xas), page); |
f3f0e1d21 khugepaged: add s... |
1652 1653 1654 1655 1656 |
VM_BUG_ON_PAGE(page_mapped(page), page); /* * The page is expected to have page_count() == 3: * - we hold a pin on it; |
77da9389b mm: Convert colla... |
1657 |
* - one reference from page cache; |
f3f0e1d21 khugepaged: add s... |
1658 1659 1660 1661 |
* - one from isolate_lru_page; */ if (!page_ref_freeze(page, 3)) { result = SCAN_PAGE_COUNT; |
042a30824 mm/khugepaged: mi... |
1662 1663 1664 |
xas_unlock_irq(&xas); putback_lru_page(page); goto out_unlock; |
f3f0e1d21 khugepaged: add s... |
1665 1666 1667 1668 1669 1670 1671 1672 1673 |
} /* * Add the page to the list to be able to undo the collapse if * something go wrong. */ list_add_tail(&page->lru, &pagelist); /* Finally, replace with the new page. */ |
4101196b1 mm: page cache: s... |
1674 |
xas_store(&xas, new_page); |
f3f0e1d21 khugepaged: add s... |
1675 |
continue; |
f3f0e1d21 khugepaged: add s... |
1676 1677 1678 |
out_unlock: unlock_page(page); put_page(page); |
042a30824 mm/khugepaged: mi... |
1679 |
goto xa_unlocked; |
f3f0e1d21 khugepaged: add s... |
1680 |
} |
99cb0dbd4 mm,thp: add read-... |
1681 1682 |
if (is_shmem) __inc_node_page_state(new_page, NR_SHMEM_THPS); |
09d91cda0 mm,thp: avoid wri... |
1683 |
else { |
99cb0dbd4 mm,thp: add read-... |
1684 |
__inc_node_page_state(new_page, NR_FILE_THPS); |
09d91cda0 mm,thp: avoid wri... |
1685 1686 |
filemap_nr_thps_inc(mapping); } |
99cb0dbd4 mm,thp: add read-... |
1687 |
|
042a30824 mm/khugepaged: mi... |
1688 1689 1690 1691 |
if (nr_none) { struct zone *zone = page_zone(new_page); __mod_node_page_state(zone->zone_pgdat, NR_FILE_PAGES, nr_none); |
99cb0dbd4 mm,thp: add read-... |
1692 1693 1694 |
if (is_shmem) __mod_node_page_state(zone->zone_pgdat, NR_SHMEM, nr_none); |
042a30824 mm/khugepaged: mi... |
1695 1696 1697 1698 |
} xa_locked: xas_unlock_irq(&xas); |
77da9389b mm: Convert colla... |
1699 |
xa_unlocked: |
042a30824 mm/khugepaged: mi... |
1700 |
|
f3f0e1d21 khugepaged: add s... |
1701 |
if (result == SCAN_SUCCEED) { |
77da9389b mm: Convert colla... |
1702 |
struct page *page, *tmp; |
f3f0e1d21 khugepaged: add s... |
1703 1704 |
/* |
77da9389b mm: Convert colla... |
1705 1706 |
* Replacing old pages with new one has succeeded, now we * need to copy the content and free the old pages. |
f3f0e1d21 khugepaged: add s... |
1707 |
*/ |
2af8ff291 mm/khugepaged: co... |
1708 |
index = start; |
f3f0e1d21 khugepaged: add s... |
1709 |
list_for_each_entry_safe(page, tmp, &pagelist, lru) { |
2af8ff291 mm/khugepaged: co... |
1710 1711 1712 1713 |
while (index < page->index) { clear_highpage(new_page + (index % HPAGE_PMD_NR)); index++; } |
f3f0e1d21 khugepaged: add s... |
1714 1715 1716 |
copy_highpage(new_page + (page->index % HPAGE_PMD_NR), page); list_del(&page->lru); |
f3f0e1d21 khugepaged: add s... |
1717 |
page->mapping = NULL; |
042a30824 mm/khugepaged: mi... |
1718 |
page_ref_unfreeze(page, 1); |
f3f0e1d21 khugepaged: add s... |
1719 1720 |
ClearPageActive(page); ClearPageUnevictable(page); |
042a30824 mm/khugepaged: mi... |
1721 |
unlock_page(page); |
f3f0e1d21 khugepaged: add s... |
1722 |
put_page(page); |
2af8ff291 mm/khugepaged: co... |
1723 1724 1725 1726 1727 |
index++; } while (index < end) { clear_highpage(new_page + (index % HPAGE_PMD_NR)); index++; |
f3f0e1d21 khugepaged: add s... |
1728 |
} |
f3f0e1d21 khugepaged: add s... |
1729 |
SetPageUptodate(new_page); |
87c460a0b mm/khugepaged: co... |
1730 |
page_ref_add(new_page, HPAGE_PMD_NR - 1); |
f3f0e1d21 khugepaged: add s... |
1731 |
mem_cgroup_commit_charge(new_page, memcg, false, true); |
99cb0dbd4 mm,thp: add read-... |
1732 1733 1734 1735 1736 1737 1738 |
if (is_shmem) { set_page_dirty(new_page); lru_cache_add_anon(new_page); } else { lru_cache_add_file(new_page); } |
1ff9e6e17 mm: memcontrol: e... |
1739 |
count_memcg_events(memcg, THP_COLLAPSE_ALLOC, 1); |
f3f0e1d21 khugepaged: add s... |
1740 |
|
042a30824 mm/khugepaged: mi... |
1741 1742 1743 1744 |
/* * Remove pte page tables, so we can re-fault the page as huge. */ retract_page_tables(mapping, start); |
f3f0e1d21 khugepaged: add s... |
1745 |
*hpage = NULL; |
87aa75290 mm: thp: inc coun... |
1746 1747 |
khugepaged_pages_collapsed++; |
f3f0e1d21 khugepaged: add s... |
1748 |
} else { |
77da9389b mm: Convert colla... |
1749 |
struct page *page; |
aaa52e340 mm/khugepaged: fi... |
1750 |
|
77da9389b mm: Convert colla... |
1751 |
/* Something went wrong: roll back page cache changes */ |
77da9389b mm: Convert colla... |
1752 |
xas_lock_irq(&xas); |
aaa52e340 mm/khugepaged: fi... |
1753 |
mapping->nrpages -= nr_none; |
99cb0dbd4 mm,thp: add read-... |
1754 1755 1756 |
if (is_shmem) shmem_uncharge(mapping->host, nr_none); |
aaa52e340 mm/khugepaged: fi... |
1757 |
|
77da9389b mm: Convert colla... |
1758 1759 |
xas_set(&xas, start); xas_for_each(&xas, page, end - 1) { |
f3f0e1d21 khugepaged: add s... |
1760 1761 |
page = list_first_entry_or_null(&pagelist, struct page, lru); |
77da9389b mm: Convert colla... |
1762 |
if (!page || xas.xa_index < page->index) { |
f3f0e1d21 khugepaged: add s... |
1763 1764 |
if (!nr_none) break; |
f3f0e1d21 khugepaged: add s... |
1765 |
nr_none--; |
59749e6ce mm: khugepaged: f... |
1766 |
/* Put holes back where they were */ |
77da9389b mm: Convert colla... |
1767 |
xas_store(&xas, NULL); |
f3f0e1d21 khugepaged: add s... |
1768 1769 |
continue; } |
77da9389b mm: Convert colla... |
1770 |
VM_BUG_ON_PAGE(page->index != xas.xa_index, page); |
f3f0e1d21 khugepaged: add s... |
1771 1772 1773 1774 |
/* Unfreeze the page. */ list_del(&page->lru); page_ref_unfreeze(page, 2); |
77da9389b mm: Convert colla... |
1775 1776 1777 |
xas_store(&xas, page); xas_pause(&xas); xas_unlock_irq(&xas); |
f3f0e1d21 khugepaged: add s... |
1778 |
unlock_page(page); |
042a30824 mm/khugepaged: mi... |
1779 |
putback_lru_page(page); |
77da9389b mm: Convert colla... |
1780 |
xas_lock_irq(&xas); |
f3f0e1d21 khugepaged: add s... |
1781 1782 |
} VM_BUG_ON(nr_none); |
77da9389b mm: Convert colla... |
1783 |
xas_unlock_irq(&xas); |
f3f0e1d21 khugepaged: add s... |
1784 |
|
f3f0e1d21 khugepaged: add s... |
1785 |
mem_cgroup_cancel_charge(new_page, memcg, true); |
f3f0e1d21 khugepaged: add s... |
1786 1787 |
new_page->mapping = NULL; } |
042a30824 mm/khugepaged: mi... |
1788 1789 |
unlock_page(new_page); |
f3f0e1d21 khugepaged: add s... |
1790 1791 1792 1793 |
out: VM_BUG_ON(!list_empty(&pagelist)); /* TODO: tracepoints */ } |
579c571e2 khugepaged: renam... |
1794 1795 |
static void khugepaged_scan_file(struct mm_struct *mm, struct file *file, pgoff_t start, struct page **hpage) |
f3f0e1d21 khugepaged: add s... |
1796 1797 |
{ struct page *page = NULL; |
579c571e2 khugepaged: renam... |
1798 |
struct address_space *mapping = file->f_mapping; |
85b392dba mm: Convert khuge... |
1799 |
XA_STATE(xas, &mapping->i_pages, start); |
f3f0e1d21 khugepaged: add s... |
1800 1801 1802 1803 1804 1805 1806 1807 |
int present, swap; int node = NUMA_NO_NODE; int result = SCAN_SUCCEED; present = 0; swap = 0; memset(khugepaged_node_load, 0, sizeof(khugepaged_node_load)); rcu_read_lock(); |
85b392dba mm: Convert khuge... |
1808 1809 |
xas_for_each(&xas, page, start + HPAGE_PMD_NR - 1) { if (xas_retry(&xas, page)) |
f3f0e1d21 khugepaged: add s... |
1810 |
continue; |
f3f0e1d21 khugepaged: add s... |
1811 |
|
85b392dba mm: Convert khuge... |
1812 |
if (xa_is_value(page)) { |
f3f0e1d21 khugepaged: add s... |
1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 |
if (++swap > khugepaged_max_ptes_swap) { result = SCAN_EXCEED_SWAP_PTE; break; } continue; } if (PageTransCompound(page)) { result = SCAN_PAGE_COMPOUND; break; } node = page_to_nid(page); if (khugepaged_scan_abort(node)) { result = SCAN_SCAN_ABORT; break; } khugepaged_node_load[node]++; if (!PageLRU(page)) { result = SCAN_PAGE_LRU; break; } |
99cb0dbd4 mm,thp: add read-... |
1836 1837 |
if (page_count(page) != 1 + page_mapcount(page) + page_has_private(page)) { |
f3f0e1d21 khugepaged: add s... |
1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 |
result = SCAN_PAGE_COUNT; break; } /* * We probably should check if the page is referenced here, but * nobody would transfer pte_young() to PageReferenced() for us. * And rmap walk here is just too costly... */ present++; if (need_resched()) { |
85b392dba mm: Convert khuge... |
1851 |
xas_pause(&xas); |
f3f0e1d21 khugepaged: add s... |
1852 |
cond_resched_rcu(); |
f3f0e1d21 khugepaged: add s... |
1853 1854 1855 1856 1857 1858 1859 1860 1861 |
} } rcu_read_unlock(); if (result == SCAN_SUCCEED) { if (present < HPAGE_PMD_NR - khugepaged_max_ptes_none) { result = SCAN_EXCEED_NONE_PTE; } else { node = khugepaged_find_target_node(); |
579c571e2 khugepaged: renam... |
1862 |
collapse_file(mm, file, start, hpage, node); |
f3f0e1d21 khugepaged: add s... |
1863 1864 1865 1866 1867 1868 |
} } /* TODO: tracepoints */ } #else |
579c571e2 khugepaged: renam... |
1869 1870 |
static void khugepaged_scan_file(struct mm_struct *mm, struct file *file, pgoff_t start, struct page **hpage) |
f3f0e1d21 khugepaged: add s... |
1871 1872 1873 |
{ BUILD_BUG(); } |
27e1f8273 khugepaged: enabl... |
1874 1875 1876 1877 1878 |
static int khugepaged_collapse_pte_mapped_thps(struct mm_slot *mm_slot) { return 0; } |
f3f0e1d21 khugepaged: add s... |
1879 |
#endif |
b46e756f5 thp: extract khug... |
1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 |
static unsigned int khugepaged_scan_mm_slot(unsigned int pages, struct page **hpage) __releases(&khugepaged_mm_lock) __acquires(&khugepaged_mm_lock) { struct mm_slot *mm_slot; struct mm_struct *mm; struct vm_area_struct *vma; int progress = 0; VM_BUG_ON(!pages); |
35f3aa39f mm: Replace spin_... |
1891 |
lockdep_assert_held(&khugepaged_mm_lock); |
b46e756f5 thp: extract khug... |
1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 |
if (khugepaged_scan.mm_slot) mm_slot = khugepaged_scan.mm_slot; else { mm_slot = list_entry(khugepaged_scan.mm_head.next, struct mm_slot, mm_node); khugepaged_scan.address = 0; khugepaged_scan.mm_slot = mm_slot; } spin_unlock(&khugepaged_mm_lock); |
27e1f8273 khugepaged: enabl... |
1902 |
khugepaged_collapse_pte_mapped_thps(mm_slot); |
b46e756f5 thp: extract khug... |
1903 1904 |
mm = mm_slot->mm; |
3b454ad35 mm: thp: use down... |
1905 1906 1907 1908 1909 1910 1911 1912 |
/* * Don't wait for semaphore (to avoid long wait times). Just move to * the next mm on the list. */ vma = NULL; if (unlikely(!down_read_trylock(&mm->mmap_sem))) goto breakouterloop_mmap_sem; if (likely(!khugepaged_test_exit(mm))) |
b46e756f5 thp: extract khug... |
1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 |
vma = find_vma(mm, khugepaged_scan.address); progress++; for (; vma; vma = vma->vm_next) { unsigned long hstart, hend; cond_resched(); if (unlikely(khugepaged_test_exit(mm))) { progress++; break; } |
50f8b92f2 mm: thp: pass cor... |
1924 |
if (!hugepage_vma_check(vma, vma->vm_flags)) { |
b46e756f5 thp: extract khug... |
1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 |
skip: progress++; continue; } hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; hend = vma->vm_end & HPAGE_PMD_MASK; if (hstart >= hend) goto skip; if (khugepaged_scan.address > hend) goto skip; if (khugepaged_scan.address < hstart) khugepaged_scan.address = hstart; VM_BUG_ON(khugepaged_scan.address & ~HPAGE_PMD_MASK); while (khugepaged_scan.address < hend) { int ret; cond_resched(); if (unlikely(khugepaged_test_exit(mm))) goto breakouterloop; VM_BUG_ON(khugepaged_scan.address < hstart || khugepaged_scan.address + HPAGE_PMD_SIZE > hend); |
99cb0dbd4 mm,thp: add read-... |
1948 |
if (IS_ENABLED(CONFIG_SHMEM) && vma->vm_file) { |
e496cf3d7 thp: introduce CO... |
1949 |
struct file *file; |
f3f0e1d21 khugepaged: add s... |
1950 1951 |
pgoff_t pgoff = linear_page_index(vma, khugepaged_scan.address); |
99cb0dbd4 mm,thp: add read-... |
1952 1953 1954 |
if (shmem_file(vma->vm_file) && !shmem_huge_enabled(vma)) |
e496cf3d7 thp: introduce CO... |
1955 1956 |
goto skip; file = get_file(vma->vm_file); |
f3f0e1d21 khugepaged: add s... |
1957 1958 |
up_read(&mm->mmap_sem); ret = 1; |
579c571e2 khugepaged: renam... |
1959 |
khugepaged_scan_file(mm, file, pgoff, hpage); |
f3f0e1d21 khugepaged: add s... |
1960 1961 1962 1963 1964 1965 |
fput(file); } else { ret = khugepaged_scan_pmd(mm, vma, khugepaged_scan.address, hpage); } |
b46e756f5 thp: extract khug... |
1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 |
/* move to next address */ khugepaged_scan.address += HPAGE_PMD_SIZE; progress += HPAGE_PMD_NR; if (ret) /* we released mmap_sem so break loop */ goto breakouterloop_mmap_sem; if (progress >= pages) goto breakouterloop; } } breakouterloop: up_read(&mm->mmap_sem); /* exit_mmap will destroy ptes after this */ breakouterloop_mmap_sem: spin_lock(&khugepaged_mm_lock); VM_BUG_ON(khugepaged_scan.mm_slot != mm_slot); /* * Release the current mm_slot if this mm is about to die, or * if we scanned all vmas of this mm. */ if (khugepaged_test_exit(mm) || !vma) { /* * Make sure that if mm_users is reaching zero while * khugepaged runs here, khugepaged_exit will find * mm_slot not pointing to the exiting mm. */ if (mm_slot->mm_node.next != &khugepaged_scan.mm_head) { khugepaged_scan.mm_slot = list_entry( mm_slot->mm_node.next, struct mm_slot, mm_node); khugepaged_scan.address = 0; } else { khugepaged_scan.mm_slot = NULL; khugepaged_full_scans++; } collect_mm_slot(mm_slot); } return progress; } static int khugepaged_has_work(void) { return !list_empty(&khugepaged_scan.mm_head) && khugepaged_enabled(); } static int khugepaged_wait_event(void) { return !list_empty(&khugepaged_scan.mm_head) || kthread_should_stop(); } static void khugepaged_do_scan(void) { struct page *hpage = NULL; unsigned int progress = 0, pass_through_head = 0; unsigned int pages = khugepaged_pages_to_scan; bool wait = true; barrier(); /* write khugepaged_pages_to_scan to local stack */ while (progress < pages) { if (!khugepaged_prealloc_page(&hpage, &wait)) break; cond_resched(); if (unlikely(kthread_should_stop() || try_to_freeze())) break; spin_lock(&khugepaged_mm_lock); if (!khugepaged_scan.mm_slot) pass_through_head++; if (khugepaged_has_work() && pass_through_head < 2) progress += khugepaged_scan_mm_slot(pages - progress, &hpage); else progress = pages; spin_unlock(&khugepaged_mm_lock); } if (!IS_ERR_OR_NULL(hpage)) put_page(hpage); } static bool khugepaged_should_wakeup(void) { return kthread_should_stop() || time_after_eq(jiffies, khugepaged_sleep_expire); } static void khugepaged_wait_work(void) { if (khugepaged_has_work()) { const unsigned long scan_sleep_jiffies = msecs_to_jiffies(khugepaged_scan_sleep_millisecs); if (!scan_sleep_jiffies) return; khugepaged_sleep_expire = jiffies + scan_sleep_jiffies; wait_event_freezable_timeout(khugepaged_wait, khugepaged_should_wakeup(), scan_sleep_jiffies); return; } if (khugepaged_enabled()) wait_event_freezable(khugepaged_wait, khugepaged_wait_event()); } static int khugepaged(void *none) { struct mm_slot *mm_slot; set_freezable(); set_user_nice(current, MAX_NICE); while (!kthread_should_stop()) { khugepaged_do_scan(); khugepaged_wait_work(); } spin_lock(&khugepaged_mm_lock); mm_slot = khugepaged_scan.mm_slot; khugepaged_scan.mm_slot = NULL; if (mm_slot) collect_mm_slot(mm_slot); spin_unlock(&khugepaged_mm_lock); return 0; } static void set_recommended_min_free_kbytes(void) { struct zone *zone; int nr_zones = 0; unsigned long recommended_min; |
b7d349c74 mm/thp: don't cou... |
2106 2107 2108 2109 2110 2111 2112 |
for_each_populated_zone(zone) { /* * We don't need to worry about fragmentation of * ZONE_MOVABLE since it only has movable pages. */ if (zone_idx(zone) > gfp_zone(GFP_USER)) continue; |
b46e756f5 thp: extract khug... |
2113 |
nr_zones++; |
b7d349c74 mm/thp: don't cou... |
2114 |
} |
b46e756f5 thp: extract khug... |
2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 |
/* Ensure 2 pageblocks are free to assist fragmentation avoidance */ recommended_min = pageblock_nr_pages * nr_zones * 2; /* * Make sure that on average at least two pageblocks are almost free * of another type, one for a migratetype to fall back to and a * second to avoid subsequent fallbacks of other types There are 3 * MIGRATE_TYPES we care about. */ recommended_min += pageblock_nr_pages * nr_zones * MIGRATE_PCPTYPES * MIGRATE_PCPTYPES; /* don't ever allow to reserve more than 5% of the lowmem */ recommended_min = min(recommended_min, (unsigned long) nr_free_buffer_pages() / 20); recommended_min <<= (PAGE_SHIFT-10); if (recommended_min > min_free_kbytes) { if (user_min_free_kbytes >= 0) pr_info("raising min_free_kbytes from %d to %lu to help transparent hugepage allocations ", min_free_kbytes, recommended_min); min_free_kbytes = recommended_min; } setup_per_zone_wmarks(); } int start_stop_khugepaged(void) { static struct task_struct *khugepaged_thread __read_mostly; static DEFINE_MUTEX(khugepaged_mutex); int err = 0; mutex_lock(&khugepaged_mutex); if (khugepaged_enabled()) { if (!khugepaged_thread) khugepaged_thread = kthread_run(khugepaged, NULL, "khugepaged"); if (IS_ERR(khugepaged_thread)) { pr_err("khugepaged: kthread_run(khugepaged) failed "); err = PTR_ERR(khugepaged_thread); khugepaged_thread = NULL; goto fail; } if (!list_empty(&khugepaged_scan.mm_head)) wake_up_interruptible(&khugepaged_wait); set_recommended_min_free_kbytes(); } else if (khugepaged_thread) { kthread_stop(khugepaged_thread); khugepaged_thread = NULL; } fail: mutex_unlock(&khugepaged_mutex); return err; } |