Blame view
mm/page_alloc.c
172 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
/* * linux/mm/page_alloc.c * * Manages the free list, the system allocates free pages here. * Note that kmalloc() lives in slab.c * * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds * Swap reorganised 29.12.95, Stephen Tweedie * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999 * Reshaped it to be a zoned allocator, Ingo Molnar, Red Hat, 1999 * Discontiguous memory support, Kanoj Sarcar, SGI, Nov 1999 * Zone balancing, Kanoj Sarcar, SGI, Jan 2000 * Per cpu hot/cold page lists, bulk allocation, Martin J. Bligh, Sept 2002 * (lots of bits borrowed from Ingo Molnar & Andrew Morton) */ |
1da177e4c Linux-2.6.12-rc2 |
16 17 18 19 20 |
#include <linux/stddef.h> #include <linux/mm.h> #include <linux/swap.h> #include <linux/interrupt.h> #include <linux/pagemap.h> |
10ed273f5 zlc_setup(): hand... |
21 |
#include <linux/jiffies.h> |
1da177e4c Linux-2.6.12-rc2 |
22 |
#include <linux/bootmem.h> |
edbe7d23b memblock: Add fin... |
23 |
#include <linux/memblock.h> |
1da177e4c Linux-2.6.12-rc2 |
24 |
#include <linux/compiler.h> |
9f1583339 [PATCH] use add_t... |
25 |
#include <linux/kernel.h> |
b1eeab676 kmemcheck: add ho... |
26 |
#include <linux/kmemcheck.h> |
1da177e4c Linux-2.6.12-rc2 |
27 28 29 30 31 |
#include <linux/module.h> #include <linux/suspend.h> #include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/slab.h> |
a238ab5b0 mm: break out pag... |
32 |
#include <linux/ratelimit.h> |
5a3135c2e oom: move prototy... |
33 |
#include <linux/oom.h> |
1da177e4c Linux-2.6.12-rc2 |
34 35 36 37 38 |
#include <linux/notifier.h> #include <linux/topology.h> #include <linux/sysctl.h> #include <linux/cpu.h> #include <linux/cpuset.h> |
bdc8cb984 [PATCH] memory ho... |
39 |
#include <linux/memory_hotplug.h> |
1da177e4c Linux-2.6.12-rc2 |
40 41 |
#include <linux/nodemask.h> #include <linux/vmalloc.h> |
a6cccdc36 mm, mem-hotplug: ... |
42 |
#include <linux/vmstat.h> |
4be38e351 [PATCH] mm: move ... |
43 |
#include <linux/mempolicy.h> |
6811378e7 [PATCH] wait_tabl... |
44 |
#include <linux/stop_machine.h> |
c713216de [PATCH] Introduce... |
45 46 |
#include <linux/sort.h> #include <linux/pfn.h> |
3fcfab16c [PATCH] separate ... |
47 |
#include <linux/backing-dev.h> |
933e312e7 [PATCH] fault-inj... |
48 |
#include <linux/fault-inject.h> |
a5d76b54a memory unplug: pa... |
49 |
#include <linux/page-isolation.h> |
52d4b9ac0 memcg: allocate a... |
50 |
#include <linux/page_cgroup.h> |
3ac7fe5a4 infrastructure to... |
51 |
#include <linux/debugobjects.h> |
dbb1f81ca kmemleak: Add kme... |
52 |
#include <linux/kmemleak.h> |
56de7263f mm: compaction: d... |
53 |
#include <linux/compaction.h> |
0d3d062a6 tracing, page-all... |
54 |
#include <trace/events/kmem.h> |
718a38211 mm: introduce dum... |
55 |
#include <linux/ftrace_event.h> |
f212ad7cf memcg: add memcg ... |
56 |
#include <linux/memcontrol.h> |
268bb0ce3 sanitize <linux/p... |
57 |
#include <linux/prefetch.h> |
041d3a8cd mm: page_alloc: i... |
58 |
#include <linux/migrate.h> |
c0a32fc5a mm: more intensiv... |
59 |
#include <linux/page-debug-flags.h> |
949f7ec57 mm, hugetlb: incl... |
60 |
#include <linux/hugetlb.h> |
8bd75c77b sched/rt: Move rt... |
61 |
#include <linux/sched/rt.h> |
1da177e4c Linux-2.6.12-rc2 |
62 63 |
#include <asm/tlbflush.h> |
ac924c603 [PATCH] setup_per... |
64 |
#include <asm/div64.h> |
1da177e4c Linux-2.6.12-rc2 |
65 |
#include "internal.h" |
728120192 numa: add generic... |
66 67 68 69 |
#ifdef CONFIG_USE_PERCPU_NUMA_NODE_ID DEFINE_PER_CPU(int, numa_node); EXPORT_PER_CPU_SYMBOL(numa_node); #endif |
7aac78988 numa: introduce n... |
70 71 72 73 74 75 76 77 78 79 |
#ifdef CONFIG_HAVE_MEMORYLESS_NODES /* * N.B., Do NOT reference the '_numa_mem_' per cpu variable directly. * It will not be defined when CONFIG_HAVE_MEMORYLESS_NODES is not defined. * Use the accessor functions set_numa_mem(), numa_mem_id() and cpu_to_mem() * defined in <linux/topology.h>. */ DEFINE_PER_CPU(int, _numa_mem_); /* Kernel "local memory" node */ EXPORT_PER_CPU_SYMBOL(_numa_mem_); #endif |
1da177e4c Linux-2.6.12-rc2 |
80 |
/* |
138089107 Memoryless nodes:... |
81 |
* Array of node states. |
1da177e4c Linux-2.6.12-rc2 |
82 |
*/ |
138089107 Memoryless nodes:... |
83 84 85 86 87 88 89 90 |
nodemask_t node_states[NR_NODE_STATES] __read_mostly = { [N_POSSIBLE] = NODE_MASK_ALL, [N_ONLINE] = { { [0] = 1UL } }, #ifndef CONFIG_NUMA [N_NORMAL_MEMORY] = { { [0] = 1UL } }, #ifdef CONFIG_HIGHMEM [N_HIGH_MEMORY] = { { [0] = 1UL } }, #endif |
20b2f52b7 numa: add CONFIG_... |
91 92 93 |
#ifdef CONFIG_MOVABLE_NODE [N_MEMORY] = { { [0] = 1UL } }, #endif |
138089107 Memoryless nodes:... |
94 95 96 97 |
[N_CPU] = { { [0] = 1UL } }, #endif /* NUMA */ }; EXPORT_SYMBOL(node_states); |
6c231b7ba [PATCH] Additions... |
98 |
unsigned long totalram_pages __read_mostly; |
cb45b0e96 [PATCH] overcommi... |
99 |
unsigned long totalreserve_pages __read_mostly; |
ab8fabd46 mm: exclude reser... |
100 101 102 103 104 105 106 |
/* * When calculating the number of globally allowed dirty pages, there * is a certain number of per-zone reserves that should not be * considered dirtyable memory. This is the sum of those reserves * over all existing zones that contribute dirtyable memory. */ unsigned long dirty_balance_reserve __read_mostly; |
1b76b02f1 mm: raise MemFree... |
107 |
int percpu_pagelist_fraction; |
dcce284a2 mm: Extend gfp ma... |
108 |
gfp_t gfp_allowed_mask __read_mostly = GFP_BOOT_MASK; |
1da177e4c Linux-2.6.12-rc2 |
109 |
|
452aa6999 mm/pm: force GFP_... |
110 111 112 113 114 115 116 117 118 |
#ifdef CONFIG_PM_SLEEP /* * The following functions are used by the suspend/hibernate code to temporarily * change gfp_allowed_mask in order to avoid using I/O during memory allocations * while devices are suspended. To avoid races with the suspend/hibernate code, * they should always be called with pm_mutex held (gfp_allowed_mask also should * only be modified with pm_mutex held, unless the suspend/hibernate code is * guaranteed not to run in parallel with that modification). */ |
c9e664f1f PM / Hibernate: F... |
119 120 121 122 |
static gfp_t saved_gfp_mask; void pm_restore_gfp_mask(void) |
452aa6999 mm/pm: force GFP_... |
123 124 |
{ WARN_ON(!mutex_is_locked(&pm_mutex)); |
c9e664f1f PM / Hibernate: F... |
125 126 127 128 |
if (saved_gfp_mask) { gfp_allowed_mask = saved_gfp_mask; saved_gfp_mask = 0; } |
452aa6999 mm/pm: force GFP_... |
129 |
} |
c9e664f1f PM / Hibernate: F... |
130 |
void pm_restrict_gfp_mask(void) |
452aa6999 mm/pm: force GFP_... |
131 |
{ |
452aa6999 mm/pm: force GFP_... |
132 |
WARN_ON(!mutex_is_locked(&pm_mutex)); |
c9e664f1f PM / Hibernate: F... |
133 134 135 |
WARN_ON(saved_gfp_mask); saved_gfp_mask = gfp_allowed_mask; gfp_allowed_mask &= ~GFP_IOFS; |
452aa6999 mm/pm: force GFP_... |
136 |
} |
f90ac3982 mm: avoid liveloc... |
137 138 139 140 141 142 143 |
bool pm_suspended_storage(void) { if ((gfp_allowed_mask & GFP_IOFS) == GFP_IOFS) return false; return true; } |
452aa6999 mm/pm: force GFP_... |
144 |
#endif /* CONFIG_PM_SLEEP */ |
d9c234005 Do not depend on ... |
145 146 147 |
#ifdef CONFIG_HUGETLB_PAGE_SIZE_VARIABLE int pageblock_order __read_mostly; #endif |
d98c7a098 [PATCH] compound ... |
148 |
static void __free_pages_ok(struct page *page, unsigned int order); |
a226f6c89 [PATCH] FRV: Clea... |
149 |
|
1da177e4c Linux-2.6.12-rc2 |
150 151 152 153 154 155 156 |
/* * results with 256, 32 in the lowmem_reserve sysctl: * 1G machine -> (16M dma, 800M-16M normal, 1G-800M high) * 1G machine -> (16M dma, 784M normal, 224M high) * NORMAL allocation will leave 784M/256 of ram reserved in the ZONE_DMA * HIGHMEM allocation will leave 224M/32 of ram reserved in ZONE_NORMAL * HIGHMEM allocation will (224M+784M)/256 of ram reserved in ZONE_DMA |
a2f1b4249 [PATCH] x86_64: A... |
157 158 159 |
* * TBD: should special case ZONE_DMA32 machines here - in those we normally * don't need any ZONE_NORMAL reservation |
1da177e4c Linux-2.6.12-rc2 |
160 |
*/ |
2f1b62486 [PATCH] reduce MA... |
161 |
int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = { |
4b51d6698 [PATCH] optional ... |
162 |
#ifdef CONFIG_ZONE_DMA |
2f1b62486 [PATCH] reduce MA... |
163 |
256, |
4b51d6698 [PATCH] optional ... |
164 |
#endif |
fb0e7942b [PATCH] reduce MA... |
165 |
#ifdef CONFIG_ZONE_DMA32 |
2f1b62486 [PATCH] reduce MA... |
166 |
256, |
fb0e7942b [PATCH] reduce MA... |
167 |
#endif |
e53ef38d0 [PATCH] reduce MA... |
168 |
#ifdef CONFIG_HIGHMEM |
2a1e274ac Create the ZONE_M... |
169 |
32, |
e53ef38d0 [PATCH] reduce MA... |
170 |
#endif |
2a1e274ac Create the ZONE_M... |
171 |
32, |
2f1b62486 [PATCH] reduce MA... |
172 |
}; |
1da177e4c Linux-2.6.12-rc2 |
173 174 |
EXPORT_SYMBOL(totalram_pages); |
1da177e4c Linux-2.6.12-rc2 |
175 |
|
15ad7cdcf [PATCH] struct se... |
176 |
static char * const zone_names[MAX_NR_ZONES] = { |
4b51d6698 [PATCH] optional ... |
177 |
#ifdef CONFIG_ZONE_DMA |
2f1b62486 [PATCH] reduce MA... |
178 |
"DMA", |
4b51d6698 [PATCH] optional ... |
179 |
#endif |
fb0e7942b [PATCH] reduce MA... |
180 |
#ifdef CONFIG_ZONE_DMA32 |
2f1b62486 [PATCH] reduce MA... |
181 |
"DMA32", |
fb0e7942b [PATCH] reduce MA... |
182 |
#endif |
2f1b62486 [PATCH] reduce MA... |
183 |
"Normal", |
e53ef38d0 [PATCH] reduce MA... |
184 |
#ifdef CONFIG_HIGHMEM |
2a1e274ac Create the ZONE_M... |
185 |
"HighMem", |
e53ef38d0 [PATCH] reduce MA... |
186 |
#endif |
2a1e274ac Create the ZONE_M... |
187 |
"Movable", |
2f1b62486 [PATCH] reduce MA... |
188 |
}; |
1da177e4c Linux-2.6.12-rc2 |
189 |
int min_free_kbytes = 1024; |
2c85f51d2 mm: also use allo... |
190 191 |
static unsigned long __meminitdata nr_kernel_pages; static unsigned long __meminitdata nr_all_pages; |
a3142c8e1 Fix section misma... |
192 |
static unsigned long __meminitdata dma_reserve; |
1da177e4c Linux-2.6.12-rc2 |
193 |
|
0ee332c14 memblock: Kill ea... |
194 195 196 197 198 199 200 201 202 203 204 |
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP static unsigned long __meminitdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES]; static unsigned long __meminitdata arch_zone_highest_possible_pfn[MAX_NR_ZONES]; static unsigned long __initdata required_kernelcore; static unsigned long __initdata required_movablecore; static unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES]; /* movable_zone is the "real" zone pages in ZONE_MOVABLE are taken from */ int movable_zone; EXPORT_SYMBOL(movable_zone); #endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ |
c713216de [PATCH] Introduce... |
205 |
|
418508c13 fix unused setup_... |
206 207 |
#if MAX_NUMNODES > 1 int nr_node_ids __read_mostly = MAX_NUMNODES; |
62bc62a87 page allocator: u... |
208 |
int nr_online_nodes __read_mostly = 1; |
418508c13 fix unused setup_... |
209 |
EXPORT_SYMBOL(nr_node_ids); |
62bc62a87 page allocator: u... |
210 |
EXPORT_SYMBOL(nr_online_nodes); |
418508c13 fix unused setup_... |
211 |
#endif |
9ef9acb05 Do not group page... |
212 |
int page_group_by_mobility_disabled __read_mostly; |
ee6f509c3 mm: factor out me... |
213 |
void set_pageblock_migratetype(struct page *page, int migratetype) |
b2a0ac887 Split the free li... |
214 |
{ |
49255c619 page allocator: m... |
215 216 217 |
if (unlikely(page_group_by_mobility_disabled)) migratetype = MIGRATE_UNMOVABLE; |
b2a0ac887 Split the free li... |
218 219 220 |
set_pageblock_flags_group(page, (unsigned long)migratetype, PB_migrate, PB_migrate_end); } |
7f33d49a2 mm, PM/Freezer: D... |
221 |
bool oom_killer_disabled __read_mostly; |
13e7444b0 [PATCH] mm: remov... |
222 |
#ifdef CONFIG_DEBUG_VM |
c6a57e19e [PATCH] memory ho... |
223 |
static int page_outside_zone_boundaries(struct zone *zone, struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
224 |
{ |
bdc8cb984 [PATCH] memory ho... |
225 226 227 |
int ret = 0; unsigned seq; unsigned long pfn = page_to_pfn(page); |
b5e6a5a27 mm/page_alloc: ad... |
228 |
unsigned long sp, start_pfn; |
c6a57e19e [PATCH] memory ho... |
229 |
|
bdc8cb984 [PATCH] memory ho... |
230 231 |
do { seq = zone_span_seqbegin(zone); |
b5e6a5a27 mm/page_alloc: ad... |
232 233 |
start_pfn = zone->zone_start_pfn; sp = zone->spanned_pages; |
108bcc96e mm: add & use zon... |
234 |
if (!zone_spans_pfn(zone, pfn)) |
bdc8cb984 [PATCH] memory ho... |
235 236 |
ret = 1; } while (zone_span_seqretry(zone, seq)); |
b5e6a5a27 mm/page_alloc: ad... |
237 238 239 240 |
if (ret) pr_err("page %lu outside zone [ %lu - %lu ] ", pfn, start_pfn, start_pfn + sp); |
bdc8cb984 [PATCH] memory ho... |
241 |
return ret; |
c6a57e19e [PATCH] memory ho... |
242 243 244 245 |
} static int page_is_consistent(struct zone *zone, struct page *page) { |
14e072984 add pfn_valid_wit... |
246 |
if (!pfn_valid_within(page_to_pfn(page))) |
c6a57e19e [PATCH] memory ho... |
247 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
248 |
if (zone != page_zone(page)) |
c6a57e19e [PATCH] memory ho... |
249 250 251 252 253 254 255 256 257 258 |
return 0; return 1; } /* * Temporary debugging check for pages not lying within a given zone. */ static int bad_range(struct zone *zone, struct page *page) { if (page_outside_zone_boundaries(zone, page)) |
1da177e4c Linux-2.6.12-rc2 |
259 |
return 1; |
c6a57e19e [PATCH] memory ho... |
260 261 |
if (!page_is_consistent(zone, page)) return 1; |
1da177e4c Linux-2.6.12-rc2 |
262 263 |
return 0; } |
13e7444b0 [PATCH] mm: remov... |
264 265 266 267 268 269 |
#else static inline int bad_range(struct zone *zone, struct page *page) { return 0; } #endif |
224abf92b [PATCH] mm: bad_p... |
270 |
static void bad_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
271 |
{ |
d936cf9b3 badpage: ratelimi... |
272 273 274 |
static unsigned long resume; static unsigned long nr_shown; static unsigned long nr_unshown; |
2a7684a23 HWPOISON: check a... |
275 276 |
/* Don't complain about poisoned pages */ if (PageHWPoison(page)) { |
22b751c3d mm: rename page s... |
277 |
page_mapcount_reset(page); /* remove PageBuddy */ |
2a7684a23 HWPOISON: check a... |
278 279 |
return; } |
d936cf9b3 badpage: ratelimi... |
280 281 282 283 284 285 286 287 288 289 |
/* * Allow a burst of 60 reports, then keep quiet for that minute; * or allow a steady drip of one report per second. */ if (nr_shown == 60) { if (time_before(jiffies, resume)) { nr_unshown++; goto out; } if (nr_unshown) { |
1e9e63650 badpage: KERN_ALE... |
290 291 292 |
printk(KERN_ALERT "BUG: Bad page state: %lu messages suppressed ", |
d936cf9b3 badpage: ratelimi... |
293 294 295 296 297 298 299 |
nr_unshown); nr_unshown = 0; } nr_shown = 0; } if (nr_shown++ == 0) resume = jiffies + 60 * HZ; |
1e9e63650 badpage: KERN_ALE... |
300 301 |
printk(KERN_ALERT "BUG: Bad page state in process %s pfn:%05lx ", |
3dc147414 badpage: replace ... |
302 |
current->comm, page_to_pfn(page)); |
718a38211 mm: introduce dum... |
303 |
dump_page(page); |
3dc147414 badpage: replace ... |
304 |
|
4f31888c1 mm: output a list... |
305 |
print_modules(); |
1da177e4c Linux-2.6.12-rc2 |
306 |
dump_stack(); |
d936cf9b3 badpage: ratelimi... |
307 |
out: |
8cc3b3922 badpage: keep any... |
308 |
/* Leave bad fields for debug, except PageBuddy could make trouble */ |
22b751c3d mm: rename page s... |
309 |
page_mapcount_reset(page); /* remove PageBuddy */ |
373d4d099 taint: add explic... |
310 |
add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE); |
1da177e4c Linux-2.6.12-rc2 |
311 |
} |
1da177e4c Linux-2.6.12-rc2 |
312 313 314 315 316 317 318 |
/* * Higher-order pages are called "compound pages". They are structured thusly: * * The first PAGE_SIZE page is called the "head page". * * The remaining PAGE_SIZE pages are called "tail pages". * |
6416b9fa4 mm: cleanup the c... |
319 320 |
* All pages have PG_compound set. All tail pages have their ->first_page * pointing at the head page. |
1da177e4c Linux-2.6.12-rc2 |
321 |
* |
41d78ba55 [PATCH] compound ... |
322 323 324 |
* The first tail page's ->lru.next holds the address of the compound page's * put_page() function. Its ->lru.prev holds the order of allocation. * This usage means that zero-order pages may not be compound. |
1da177e4c Linux-2.6.12-rc2 |
325 |
*/ |
d98c7a098 [PATCH] compound ... |
326 327 328 |
static void free_compound_page(struct page *page) { |
d85f33855 Make page->privat... |
329 |
__free_pages_ok(page, compound_order(page)); |
d98c7a098 [PATCH] compound ... |
330 |
} |
01ad1c082 mm: export prep_c... |
331 |
void prep_compound_page(struct page *page, unsigned long order) |
1da177e4c Linux-2.6.12-rc2 |
332 333 334 |
{ int i; int nr_pages = 1 << order; |
18229df5b hugetlb: pull gig... |
335 336 337 338 339 340 |
set_compound_page_dtor(page, free_compound_page); set_compound_order(page, order); __SetPageHead(page); for (i = 1; i < nr_pages; i++) { struct page *p = page + i; |
18229df5b hugetlb: pull gig... |
341 |
__SetPageTail(p); |
58a84aa92 thp: set compound... |
342 |
set_page_count(p, 0); |
18229df5b hugetlb: pull gig... |
343 344 345 |
p->first_page = page; } } |
59ff42163 thp: comment remi... |
346 |
/* update __split_huge_page_refcount if you change this function */ |
8cc3b3922 badpage: keep any... |
347 |
static int destroy_compound_page(struct page *page, unsigned long order) |
1da177e4c Linux-2.6.12-rc2 |
348 349 350 |
{ int i; int nr_pages = 1 << order; |
8cc3b3922 badpage: keep any... |
351 |
int bad = 0; |
1da177e4c Linux-2.6.12-rc2 |
352 |
|
0bb2c7637 mm/page_alloc.c: ... |
353 |
if (unlikely(compound_order(page) != order)) { |
224abf92b [PATCH] mm: bad_p... |
354 |
bad_page(page); |
8cc3b3922 badpage: keep any... |
355 356 |
bad++; } |
1da177e4c Linux-2.6.12-rc2 |
357 |
|
6d7779538 mm: optimize comp... |
358 |
__ClearPageHead(page); |
8cc3b3922 badpage: keep any... |
359 |
|
18229df5b hugetlb: pull gig... |
360 361 |
for (i = 1; i < nr_pages; i++) { struct page *p = page + i; |
1da177e4c Linux-2.6.12-rc2 |
362 |
|
e713a21d8 trivial: Fix dubi... |
363 |
if (unlikely(!PageTail(p) || (p->first_page != page))) { |
224abf92b [PATCH] mm: bad_p... |
364 |
bad_page(page); |
8cc3b3922 badpage: keep any... |
365 366 |
bad++; } |
d85f33855 Make page->privat... |
367 |
__ClearPageTail(p); |
1da177e4c Linux-2.6.12-rc2 |
368 |
} |
8cc3b3922 badpage: keep any... |
369 370 |
return bad; |
1da177e4c Linux-2.6.12-rc2 |
371 |
} |
1da177e4c Linux-2.6.12-rc2 |
372 |
|
17cf44064 [PATCH] mm: clean... |
373 374 375 |
static inline void prep_zero_page(struct page *page, int order, gfp_t gfp_flags) { int i; |
6626c5d53 [PATCH] mm: prep_... |
376 377 378 379 |
/* * clear_highpage() will use KM_USER0, so it's a bug to use __GFP_ZERO * and __GFP_HIGHMEM from hard or soft interrupt context. */ |
725d704ec [PATCH] mm: VM_BU... |
380 |
VM_BUG_ON((gfp_flags & __GFP_HIGHMEM) && in_interrupt()); |
17cf44064 [PATCH] mm: clean... |
381 382 383 |
for (i = 0; i < (1 << order); i++) clear_highpage(page + i); } |
c0a32fc5a mm: more intensiv... |
384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 |
#ifdef CONFIG_DEBUG_PAGEALLOC unsigned int _debug_guardpage_minorder; static int __init debug_guardpage_minorder_setup(char *buf) { unsigned long res; if (kstrtoul(buf, 10, &res) < 0 || res > MAX_ORDER / 2) { printk(KERN_ERR "Bad debug_guardpage_minorder value "); return 0; } _debug_guardpage_minorder = res; printk(KERN_INFO "Setting debug_guardpage_minorder to %lu ", res); return 0; } __setup("debug_guardpage_minorder=", debug_guardpage_minorder_setup); static inline void set_page_guard_flag(struct page *page) { __set_bit(PAGE_DEBUG_FLAG_GUARD, &page->debug_flags); } static inline void clear_page_guard_flag(struct page *page) { __clear_bit(PAGE_DEBUG_FLAG_GUARD, &page->debug_flags); } #else static inline void set_page_guard_flag(struct page *page) { } static inline void clear_page_guard_flag(struct page *page) { } #endif |
6aa3001b2 [PATCH] page_allo... |
416 417 |
static inline void set_page_order(struct page *page, int order) { |
4c21e2f24 [PATCH] mm: split... |
418 |
set_page_private(page, order); |
676165a8a [PATCH] Fix buddy... |
419 |
__SetPageBuddy(page); |
1da177e4c Linux-2.6.12-rc2 |
420 421 422 423 |
} static inline void rmv_page_order(struct page *page) { |
676165a8a [PATCH] Fix buddy... |
424 |
__ClearPageBuddy(page); |
4c21e2f24 [PATCH] mm: split... |
425 |
set_page_private(page, 0); |
1da177e4c Linux-2.6.12-rc2 |
426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 |
} /* * Locate the struct page for both the matching buddy in our * pair (buddy1) and the combined O(n+1) page they form (page). * * 1) Any buddy B1 will have an order O twin B2 which satisfies * the following equation: * B2 = B1 ^ (1 << O) * For example, if the starting buddy (buddy2) is #8 its order * 1 buddy is #10: * B2 = 8 ^ (1 << 1) = 8 ^ 2 = 10 * * 2) Any buddy B will have an order O+1 parent P which * satisfies the following equation: * P = B & ~(1 << O) * |
d6e05edc5 spelling fixes |
443 |
* Assumption: *_mem_map is contiguous at least up to MAX_ORDER |
1da177e4c Linux-2.6.12-rc2 |
444 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
445 |
static inline unsigned long |
43506fad2 mm/page_alloc.c: ... |
446 |
__find_buddy_index(unsigned long page_idx, unsigned int order) |
1da177e4c Linux-2.6.12-rc2 |
447 |
{ |
43506fad2 mm/page_alloc.c: ... |
448 |
return page_idx ^ (1 << order); |
1da177e4c Linux-2.6.12-rc2 |
449 450 451 452 453 |
} /* * This function checks whether a page is free && is the buddy * we can do coalesce a page and its buddy if |
13e7444b0 [PATCH] mm: remov... |
454 |
* (a) the buddy is not in a hole && |
676165a8a [PATCH] Fix buddy... |
455 |
* (b) the buddy is in the buddy system && |
cb2b95e1c [PATCH] zone hand... |
456 457 |
* (c) a page and its buddy have the same order && * (d) a page and its buddy are in the same zone. |
676165a8a [PATCH] Fix buddy... |
458 |
* |
5f24ce5fd thp: remove PG_buddy |
459 460 |
* For recording whether a page is in the buddy system, we set ->_mapcount -2. * Setting, clearing, and testing _mapcount -2 is serialized by zone->lock. |
1da177e4c Linux-2.6.12-rc2 |
461 |
* |
676165a8a [PATCH] Fix buddy... |
462 |
* For recording page's order, we use page_private(page). |
1da177e4c Linux-2.6.12-rc2 |
463 |
*/ |
cb2b95e1c [PATCH] zone hand... |
464 465 |
static inline int page_is_buddy(struct page *page, struct page *buddy, int order) |
1da177e4c Linux-2.6.12-rc2 |
466 |
{ |
14e072984 add pfn_valid_wit... |
467 |
if (!pfn_valid_within(page_to_pfn(buddy))) |
13e7444b0 [PATCH] mm: remov... |
468 |
return 0; |
13e7444b0 [PATCH] mm: remov... |
469 |
|
cb2b95e1c [PATCH] zone hand... |
470 471 |
if (page_zone_id(page) != page_zone_id(buddy)) return 0; |
c0a32fc5a mm: more intensiv... |
472 473 474 475 |
if (page_is_guard(buddy) && page_order(buddy) == order) { VM_BUG_ON(page_count(buddy) != 0); return 1; } |
cb2b95e1c [PATCH] zone hand... |
476 |
if (PageBuddy(buddy) && page_order(buddy) == order) { |
a3af9c389 page allocator: d... |
477 |
VM_BUG_ON(page_count(buddy) != 0); |
6aa3001b2 [PATCH] page_allo... |
478 |
return 1; |
676165a8a [PATCH] Fix buddy... |
479 |
} |
6aa3001b2 [PATCH] page_allo... |
480 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 |
} /* * Freeing function for a buddy system allocator. * * The concept of a buddy system is to maintain direct-mapped table * (containing bit values) for memory blocks of various "orders". * The bottom level table contains the map for the smallest allocatable * units of memory (here, pages), and each level above it describes * pairs of units from the levels below, hence, "buddies". * At a high level, all that happens here is marking the table entry * at the bottom level available, and propagating the changes upward * as necessary, plus some accounting needed to play nicely with other * parts of the VM system. * At each level, we keep a list of pages, which are heads of continuous |
5f24ce5fd thp: remove PG_buddy |
496 |
* free pages of length of (1 << order) and marked with _mapcount -2. Page's |
4c21e2f24 [PATCH] mm: split... |
497 |
* order is recorded in page_private(page) field. |
1da177e4c Linux-2.6.12-rc2 |
498 |
* So when we are allocating or freeing one, we can derive the state of the |
5f63b720b mm: page_alloc: r... |
499 500 |
* other. That is, if we allocate a small block, and both were * free, the remainder of the region must be split into blocks. |
1da177e4c Linux-2.6.12-rc2 |
501 |
* If a block is freed, and its buddy is also free, then this |
5f63b720b mm: page_alloc: r... |
502 |
* triggers coalescing into a block of larger size. |
1da177e4c Linux-2.6.12-rc2 |
503 |
* |
6d49e352a propagate name ch... |
504 |
* -- nyc |
1da177e4c Linux-2.6.12-rc2 |
505 |
*/ |
48db57f8f [PATCH] mm: free_... |
506 |
static inline void __free_one_page(struct page *page, |
ed0ae21dc page allocator: d... |
507 508 |
struct zone *zone, unsigned int order, int migratetype) |
1da177e4c Linux-2.6.12-rc2 |
509 510 |
{ unsigned long page_idx; |
6dda9d55b page allocator: r... |
511 |
unsigned long combined_idx; |
43506fad2 mm/page_alloc.c: ... |
512 |
unsigned long uninitialized_var(buddy_idx); |
6dda9d55b page allocator: r... |
513 |
struct page *buddy; |
1da177e4c Linux-2.6.12-rc2 |
514 |
|
d29bb9782 mm/page_alloc: ad... |
515 |
VM_BUG_ON(!zone_is_initialized(zone)); |
224abf92b [PATCH] mm: bad_p... |
516 |
if (unlikely(PageCompound(page))) |
8cc3b3922 badpage: keep any... |
517 518 |
if (unlikely(destroy_compound_page(page, order))) return; |
1da177e4c Linux-2.6.12-rc2 |
519 |
|
ed0ae21dc page allocator: d... |
520 |
VM_BUG_ON(migratetype == -1); |
1da177e4c Linux-2.6.12-rc2 |
521 |
page_idx = page_to_pfn(page) & ((1 << MAX_ORDER) - 1); |
f2260e6b1 page allocator: u... |
522 |
VM_BUG_ON(page_idx & ((1 << order) - 1)); |
725d704ec [PATCH] mm: VM_BU... |
523 |
VM_BUG_ON(bad_range(zone, page)); |
1da177e4c Linux-2.6.12-rc2 |
524 |
|
1da177e4c Linux-2.6.12-rc2 |
525 |
while (order < MAX_ORDER-1) { |
43506fad2 mm/page_alloc.c: ... |
526 527 |
buddy_idx = __find_buddy_index(page_idx, order); buddy = page + (buddy_idx - page_idx); |
cb2b95e1c [PATCH] zone hand... |
528 |
if (!page_is_buddy(page, buddy, order)) |
3c82d0ce2 buddy: clarify co... |
529 |
break; |
c0a32fc5a mm: more intensiv... |
530 531 532 533 534 535 536 |
/* * Our buddy is free or it is CONFIG_DEBUG_PAGEALLOC guard page, * merge with it and move up one order. */ if (page_is_guard(buddy)) { clear_page_guard_flag(buddy); set_page_private(page, 0); |
d1ce749a0 cma: count free C... |
537 538 |
__mod_zone_freepage_state(zone, 1 << order, migratetype); |
c0a32fc5a mm: more intensiv... |
539 540 541 542 543 |
} else { list_del(&buddy->lru); zone->free_area[order].nr_free--; rmv_page_order(buddy); } |
43506fad2 mm/page_alloc.c: ... |
544 |
combined_idx = buddy_idx & page_idx; |
1da177e4c Linux-2.6.12-rc2 |
545 546 547 548 549 |
page = page + (combined_idx - page_idx); page_idx = combined_idx; order++; } set_page_order(page, order); |
6dda9d55b page allocator: r... |
550 551 552 553 554 555 556 557 558 |
/* * If this is not the largest possible page, check if the buddy * of the next-highest order is free. If it is, it's possible * that pages are being freed that will coalesce soon. In case, * that is happening, add the free page to the tail of the list * so it's less likely to be used soon and more likely to be merged * as a higher order page */ |
b7f50cfa3 mm, page-allocato... |
559 |
if ((order < MAX_ORDER-2) && pfn_valid_within(page_to_pfn(buddy))) { |
6dda9d55b page allocator: r... |
560 |
struct page *higher_page, *higher_buddy; |
43506fad2 mm/page_alloc.c: ... |
561 562 563 |
combined_idx = buddy_idx & page_idx; higher_page = page + (combined_idx - page_idx); buddy_idx = __find_buddy_index(combined_idx, order + 1); |
0ba8f2d59 mm/page_alloc: fi... |
564 |
higher_buddy = higher_page + (buddy_idx - combined_idx); |
6dda9d55b page allocator: r... |
565 566 567 568 569 570 571 572 573 |
if (page_is_buddy(higher_page, higher_buddy, order + 1)) { list_add_tail(&page->lru, &zone->free_area[order].free_list[migratetype]); goto out; } } list_add(&page->lru, &zone->free_area[order].free_list[migratetype]); out: |
1da177e4c Linux-2.6.12-rc2 |
574 575 |
zone->free_area[order].nr_free++; } |
224abf92b [PATCH] mm: bad_p... |
576 |
static inline int free_pages_check(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
577 |
{ |
92be2e33b [PATCH] mm: micro... |
578 579 |
if (unlikely(page_mapcount(page) | (page->mapping != NULL) | |
a3af9c389 page allocator: d... |
580 |
(atomic_read(&page->_count) != 0) | |
f212ad7cf memcg: add memcg ... |
581 582 |
(page->flags & PAGE_FLAGS_CHECK_AT_FREE) | (mem_cgroup_bad_page_check(page)))) { |
224abf92b [PATCH] mm: bad_p... |
583 |
bad_page(page); |
79f4b7bf3 badpage: simplify... |
584 |
return 1; |
8cc3b3922 badpage: keep any... |
585 |
} |
22b751c3d mm: rename page s... |
586 |
page_nid_reset_last(page); |
79f4b7bf3 badpage: simplify... |
587 588 589 |
if (page->flags & PAGE_FLAGS_CHECK_AT_PREP) page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP; return 0; |
1da177e4c Linux-2.6.12-rc2 |
590 591 592 |
} /* |
5f8dcc212 page-allocator: s... |
593 |
* Frees a number of pages from the PCP lists |
1da177e4c Linux-2.6.12-rc2 |
594 |
* Assumes all pages on list are in same zone, and of same order. |
207f36eec [PATCH] remove in... |
595 |
* count is the number of pages to free. |
1da177e4c Linux-2.6.12-rc2 |
596 597 598 599 600 601 602 |
* * If the zone was previously in an "all pages pinned" state then look to * see if this freeing clears that state. * * And clear the zone's pages_scanned counter, to hold off the "all pages are * pinned" detection logic. */ |
5f8dcc212 page-allocator: s... |
603 604 |
static void free_pcppages_bulk(struct zone *zone, int count, struct per_cpu_pages *pcp) |
1da177e4c Linux-2.6.12-rc2 |
605 |
{ |
5f8dcc212 page-allocator: s... |
606 |
int migratetype = 0; |
a6f9edd65 page-allocator: m... |
607 |
int batch_free = 0; |
72853e299 mm: page allocato... |
608 |
int to_free = count; |
5f8dcc212 page-allocator: s... |
609 |
|
c54ad30c7 [PATCH] mm: pagea... |
610 |
spin_lock(&zone->lock); |
93e4a89a8 mm: restore zone-... |
611 |
zone->all_unreclaimable = 0; |
1da177e4c Linux-2.6.12-rc2 |
612 |
zone->pages_scanned = 0; |
f2260e6b1 page allocator: u... |
613 |
|
72853e299 mm: page allocato... |
614 |
while (to_free) { |
48db57f8f [PATCH] mm: free_... |
615 |
struct page *page; |
5f8dcc212 page-allocator: s... |
616 617 618 |
struct list_head *list; /* |
a6f9edd65 page-allocator: m... |
619 620 621 622 623 |
* Remove pages from lists in a round-robin fashion. A * batch_free count is maintained that is incremented when an * empty list is encountered. This is so more pages are freed * off fuller lists instead of spinning excessively around empty * lists |
5f8dcc212 page-allocator: s... |
624 625 |
*/ do { |
a6f9edd65 page-allocator: m... |
626 |
batch_free++; |
5f8dcc212 page-allocator: s... |
627 628 629 630 |
if (++migratetype == MIGRATE_PCPTYPES) migratetype = 0; list = &pcp->lists[migratetype]; } while (list_empty(list)); |
48db57f8f [PATCH] mm: free_... |
631 |
|
1d16871d8 mm: batch-free pc... |
632 633 634 |
/* This is the only non-empty list. Free them all. */ if (batch_free == MIGRATE_PCPTYPES) batch_free = to_free; |
a6f9edd65 page-allocator: m... |
635 |
do { |
770c8aaaf mm: fix tracing i... |
636 |
int mt; /* migratetype of the to-be-freed page */ |
a6f9edd65 page-allocator: m... |
637 638 639 |
page = list_entry(list->prev, struct page, lru); /* must delete as __free_one_page list manipulates */ list_del(&page->lru); |
b12c4ad14 mm: page_alloc: u... |
640 |
mt = get_freepage_migratetype(page); |
a7016235a mm: fix migratety... |
641 |
/* MIGRATE_MOVABLE list may include MIGRATE_RESERVEs */ |
770c8aaaf mm: fix tracing i... |
642 643 |
__free_one_page(page, zone, 0, mt); trace_mm_page_pcpu_drain(page, 0, mt); |
194159fbc mm: remove MIGRAT... |
644 |
if (likely(!is_migrate_isolate_page(page))) { |
97d0da220 memory-hotplug: f... |
645 646 647 648 |
__mod_zone_page_state(zone, NR_FREE_PAGES, 1); if (is_migrate_cma(mt)) __mod_zone_page_state(zone, NR_FREE_CMA_PAGES, 1); } |
72853e299 mm: page allocato... |
649 |
} while (--to_free && --batch_free && !list_empty(list)); |
1da177e4c Linux-2.6.12-rc2 |
650 |
} |
c54ad30c7 [PATCH] mm: pagea... |
651 |
spin_unlock(&zone->lock); |
1da177e4c Linux-2.6.12-rc2 |
652 |
} |
ed0ae21dc page allocator: d... |
653 654 |
static void free_one_page(struct zone *zone, struct page *page, int order, int migratetype) |
1da177e4c Linux-2.6.12-rc2 |
655 |
{ |
006d22d9b [PATCH] Optimize ... |
656 |
spin_lock(&zone->lock); |
93e4a89a8 mm: restore zone-... |
657 |
zone->all_unreclaimable = 0; |
006d22d9b [PATCH] Optimize ... |
658 |
zone->pages_scanned = 0; |
f2260e6b1 page allocator: u... |
659 |
|
ed0ae21dc page allocator: d... |
660 |
__free_one_page(page, zone, order, migratetype); |
194159fbc mm: remove MIGRAT... |
661 |
if (unlikely(!is_migrate_isolate(migratetype))) |
d1ce749a0 cma: count free C... |
662 |
__mod_zone_freepage_state(zone, 1 << order, migratetype); |
006d22d9b [PATCH] Optimize ... |
663 |
spin_unlock(&zone->lock); |
48db57f8f [PATCH] mm: free_... |
664 |
} |
ec95f53aa mm: introduce fre... |
665 |
static bool free_pages_prepare(struct page *page, unsigned int order) |
48db57f8f [PATCH] mm: free_... |
666 |
{ |
1da177e4c Linux-2.6.12-rc2 |
667 |
int i; |
8cc3b3922 badpage: keep any... |
668 |
int bad = 0; |
1da177e4c Linux-2.6.12-rc2 |
669 |
|
b413d48aa mm-tracepoint: re... |
670 |
trace_mm_page_free(page, order); |
b1eeab676 kmemcheck: add ho... |
671 |
kmemcheck_free_shadow(page, order); |
8dd60a3a6 thp: clear compou... |
672 673 674 675 |
if (PageAnon(page)) page->mapping = NULL; for (i = 0; i < (1 << order); i++) bad += free_pages_check(page + i); |
8cc3b3922 badpage: keep any... |
676 |
if (bad) |
ec95f53aa mm: introduce fre... |
677 |
return false; |
689bcebfd [PATCH] unpaged: ... |
678 |
|
3ac7fe5a4 infrastructure to... |
679 |
if (!PageHighMem(page)) { |
9858db504 [PATCH] mm: locks... |
680 |
debug_check_no_locks_freed(page_address(page),PAGE_SIZE<<order); |
3ac7fe5a4 infrastructure to... |
681 682 683 |
debug_check_no_obj_freed(page_address(page), PAGE_SIZE << order); } |
dafb13673 [PATCH] mm: arch_... |
684 |
arch_free_page(page, order); |
48db57f8f [PATCH] mm: free_... |
685 |
kernel_map_pages(page, 1 << order, 0); |
dafb13673 [PATCH] mm: arch_... |
686 |
|
ec95f53aa mm: introduce fre... |
687 688 689 690 691 692 |
return true; } static void __free_pages_ok(struct page *page, unsigned int order) { unsigned long flags; |
95e344124 mm: remain migrat... |
693 |
int migratetype; |
ec95f53aa mm: introduce fre... |
694 695 696 |
if (!free_pages_prepare(page, order)) return; |
c54ad30c7 [PATCH] mm: pagea... |
697 |
local_irq_save(flags); |
f8891e5e1 [PATCH] Light wei... |
698 |
__count_vm_events(PGFREE, 1 << order); |
95e344124 mm: remain migrat... |
699 700 701 |
migratetype = get_pageblock_migratetype(page); set_freepage_migratetype(page, migratetype); free_one_page(page_zone(page), page, order, migratetype); |
c54ad30c7 [PATCH] mm: pagea... |
702 |
local_irq_restore(flags); |
1da177e4c Linux-2.6.12-rc2 |
703 |
} |
9feedc9d8 mm: introduce new... |
704 705 706 707 708 709 710 |
/* * Read access to zone->managed_pages is safe because it's unsigned long, * but we still need to serialize writers. Currently all callers of * __free_pages_bootmem() except put_page_bootmem() should only be used * at boot time. So for shorter boot time, we shift the burden to * put_page_bootmem() to serialize writers. */ |
af370fb8c memory hotplug: s... |
711 |
void __meminit __free_pages_bootmem(struct page *page, unsigned int order) |
a226f6c89 [PATCH] FRV: Clea... |
712 |
{ |
c3993076f mm: page_alloc: g... |
713 714 |
unsigned int nr_pages = 1 << order; unsigned int loop; |
a226f6c89 [PATCH] FRV: Clea... |
715 |
|
c3993076f mm: page_alloc: g... |
716 717 718 719 720 721 722 723 |
prefetchw(page); for (loop = 0; loop < nr_pages; loop++) { struct page *p = &page[loop]; if (loop + 1 < nr_pages) prefetchw(p + 1); __ClearPageReserved(p); set_page_count(p, 0); |
a226f6c89 [PATCH] FRV: Clea... |
724 |
} |
c3993076f mm: page_alloc: g... |
725 |
|
9feedc9d8 mm: introduce new... |
726 |
page_zone(page)->managed_pages += 1 << order; |
c3993076f mm: page_alloc: g... |
727 728 |
set_page_refcounted(page); __free_pages(page, order); |
a226f6c89 [PATCH] FRV: Clea... |
729 |
} |
47118af07 mm: mmzone: MIGRA... |
730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 |
#ifdef CONFIG_CMA /* Free whole pageblock and set it's migration type to MIGRATE_CMA. */ void __init init_cma_reserved_pageblock(struct page *page) { unsigned i = pageblock_nr_pages; struct page *p = page; do { __ClearPageReserved(p); set_page_count(p, 0); } while (++p, --i); set_page_refcounted(page); set_pageblock_migratetype(page, MIGRATE_CMA); __free_pages(page, pageblock_order); totalram_pages += pageblock_nr_pages; |
41a797344 mm: cma: fix acco... |
746 747 748 749 |
#ifdef CONFIG_HIGHMEM if (PageHighMem(page)) totalhigh_pages += pageblock_nr_pages; #endif |
47118af07 mm: mmzone: MIGRA... |
750 751 |
} #endif |
1da177e4c Linux-2.6.12-rc2 |
752 753 754 755 756 757 758 759 760 761 762 763 764 |
/* * The order of subdivision here is critical for the IO subsystem. * Please do not alter this order without good reasons and regression * testing. Specifically, as large blocks of memory are subdivided, * the order in which smaller blocks are delivered depends on the order * they're subdivided in this function. This is the primary factor * influencing the order in which pages are delivered to the IO * subsystem according to empirical testing, and this is also justified * by considering the behavior of a buddy system containing a single * large block of memory acted on by a series of small allocations. * This behavior is a critical factor in sglist merging's success. * |
6d49e352a propagate name ch... |
765 |
* -- nyc |
1da177e4c Linux-2.6.12-rc2 |
766 |
*/ |
085cc7d5d [PATCH] mm: page_... |
767 |
static inline void expand(struct zone *zone, struct page *page, |
b2a0ac887 Split the free li... |
768 769 |
int low, int high, struct free_area *area, int migratetype) |
1da177e4c Linux-2.6.12-rc2 |
770 771 772 773 774 775 776 |
{ unsigned long size = 1 << high; while (high > low) { area--; high--; size >>= 1; |
725d704ec [PATCH] mm: VM_BU... |
777 |
VM_BUG_ON(bad_range(zone, &page[size])); |
c0a32fc5a mm: more intensiv... |
778 779 780 781 782 783 784 785 786 787 788 789 790 |
#ifdef CONFIG_DEBUG_PAGEALLOC if (high < debug_guardpage_minorder()) { /* * Mark as guard pages (or page), that will allow to * merge back to allocator when buddy will be freed. * Corresponding page table entries will not be touched, * pages will stay not present in virtual address space */ INIT_LIST_HEAD(&page[size].lru); set_page_guard_flag(&page[size]); set_page_private(&page[size], high); /* Guard pages are not available for any usage */ |
d1ce749a0 cma: count free C... |
791 792 |
__mod_zone_freepage_state(zone, -(1 << high), migratetype); |
c0a32fc5a mm: more intensiv... |
793 794 795 |
continue; } #endif |
b2a0ac887 Split the free li... |
796 |
list_add(&page[size].lru, &area->free_list[migratetype]); |
1da177e4c Linux-2.6.12-rc2 |
797 798 799 |
area->nr_free++; set_page_order(&page[size], high); } |
1da177e4c Linux-2.6.12-rc2 |
800 |
} |
1da177e4c Linux-2.6.12-rc2 |
801 802 803 |
/* * This page is about to be returned from the page allocator */ |
2a7684a23 HWPOISON: check a... |
804 |
static inline int check_new_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
805 |
{ |
92be2e33b [PATCH] mm: micro... |
806 807 |
if (unlikely(page_mapcount(page) | (page->mapping != NULL) | |
a3af9c389 page allocator: d... |
808 |
(atomic_read(&page->_count) != 0) | |
f212ad7cf memcg: add memcg ... |
809 810 |
(page->flags & PAGE_FLAGS_CHECK_AT_PREP) | (mem_cgroup_bad_page_check(page)))) { |
224abf92b [PATCH] mm: bad_p... |
811 |
bad_page(page); |
689bcebfd [PATCH] unpaged: ... |
812 |
return 1; |
8cc3b3922 badpage: keep any... |
813 |
} |
2a7684a23 HWPOISON: check a... |
814 815 816 817 818 819 820 821 822 823 824 825 |
return 0; } static int prep_new_page(struct page *page, int order, gfp_t gfp_flags) { int i; for (i = 0; i < (1 << order); i++) { struct page *p = page + i; if (unlikely(check_new_page(p))) return 1; } |
689bcebfd [PATCH] unpaged: ... |
826 |
|
4c21e2f24 [PATCH] mm: split... |
827 |
set_page_private(page, 0); |
7835e98b2 [PATCH] remove se... |
828 |
set_page_refcounted(page); |
cc1025090 [PATCH] mm: add a... |
829 830 |
arch_alloc_page(page, order); |
1da177e4c Linux-2.6.12-rc2 |
831 |
kernel_map_pages(page, 1 << order, 1); |
17cf44064 [PATCH] mm: clean... |
832 833 834 835 836 837 |
if (gfp_flags & __GFP_ZERO) prep_zero_page(page, order, gfp_flags); if (order && (gfp_flags & __GFP_COMP)) prep_compound_page(page, order); |
689bcebfd [PATCH] unpaged: ... |
838 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
839 |
} |
56fd56b86 Bias the location... |
840 841 842 843 |
/* * Go through the free lists for the given migratetype and remove * the smallest available page from the freelists */ |
728ec980f page allocator: i... |
844 845 |
static inline struct page *__rmqueue_smallest(struct zone *zone, unsigned int order, |
56fd56b86 Bias the location... |
846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 |
int migratetype) { unsigned int current_order; struct free_area * area; struct page *page; /* Find a page of the appropriate size in the preferred list */ for (current_order = order; current_order < MAX_ORDER; ++current_order) { area = &(zone->free_area[current_order]); if (list_empty(&area->free_list[migratetype])) continue; page = list_entry(area->free_list[migratetype].next, struct page, lru); list_del(&page->lru); rmv_page_order(page); area->nr_free--; |
56fd56b86 Bias the location... |
863 864 865 866 867 868 |
expand(zone, page, order, current_order, area, migratetype); return page; } return NULL; } |
b2a0ac887 Split the free li... |
869 870 871 872 |
/* * This array describes the order lists are fallen back to when * the free lists for the desirable migrate type are depleted */ |
47118af07 mm: mmzone: MIGRA... |
873 874 875 876 877 878 879 880 881 |
static int fallbacks[MIGRATE_TYPES][4] = { [MIGRATE_UNMOVABLE] = { MIGRATE_RECLAIMABLE, MIGRATE_MOVABLE, MIGRATE_RESERVE }, [MIGRATE_RECLAIMABLE] = { MIGRATE_UNMOVABLE, MIGRATE_MOVABLE, MIGRATE_RESERVE }, #ifdef CONFIG_CMA [MIGRATE_MOVABLE] = { MIGRATE_CMA, MIGRATE_RECLAIMABLE, MIGRATE_UNMOVABLE, MIGRATE_RESERVE }, [MIGRATE_CMA] = { MIGRATE_RESERVE }, /* Never used */ #else [MIGRATE_MOVABLE] = { MIGRATE_RECLAIMABLE, MIGRATE_UNMOVABLE, MIGRATE_RESERVE }, #endif |
6d4a49160 mm: page_alloc: c... |
882 |
[MIGRATE_RESERVE] = { MIGRATE_RESERVE }, /* Never used */ |
194159fbc mm: remove MIGRAT... |
883 |
#ifdef CONFIG_MEMORY_ISOLATION |
6d4a49160 mm: page_alloc: c... |
884 |
[MIGRATE_ISOLATE] = { MIGRATE_RESERVE }, /* Never used */ |
194159fbc mm: remove MIGRAT... |
885 |
#endif |
b2a0ac887 Split the free li... |
886 |
}; |
c361be55b Move free pages b... |
887 888 |
/* * Move the free pages in a range to the free lists of the requested type. |
d9c234005 Do not depend on ... |
889 |
* Note that start_page and end_pages are not aligned on a pageblock |
c361be55b Move free pages b... |
890 891 |
* boundary. If alignment is required, use move_freepages_block() */ |
435b405c0 memory-hotplug: f... |
892 |
int move_freepages(struct zone *zone, |
b69a7288e mm/page_alloc.c: ... |
893 894 |
struct page *start_page, struct page *end_page, int migratetype) |
c361be55b Move free pages b... |
895 896 897 |
{ struct page *page; unsigned long order; |
d100313fd Fix calculation i... |
898 |
int pages_moved = 0; |
c361be55b Move free pages b... |
899 900 901 902 903 904 905 |
#ifndef CONFIG_HOLES_IN_ZONE /* * page_zone is not safe to call in this context when * CONFIG_HOLES_IN_ZONE is set. This bug check is probably redundant * anyway as we check zone boundaries in move_freepages_block(). * Remove at a later date when no bug reports exist related to |
ac0e5b7a6 remove PAGE_GROUP... |
906 |
* grouping pages by mobility |
c361be55b Move free pages b... |
907 908 909 910 911 |
*/ BUG_ON(page_zone(start_page) != page_zone(end_page)); #endif for (page = start_page; page <= end_page;) { |
344c790e3 mm: make setup_zo... |
912 913 |
/* Make sure we are not inadvertently changing nodes */ VM_BUG_ON(page_to_nid(page) != zone_to_nid(zone)); |
c361be55b Move free pages b... |
914 915 916 917 918 919 920 921 922 923 924 |
if (!pfn_valid_within(page_to_pfn(page))) { page++; continue; } if (!PageBuddy(page)) { page++; continue; } order = page_order(page); |
84be48d84 mm/page_alloc.c: ... |
925 926 |
list_move(&page->lru, &zone->free_area[order].free_list[migratetype]); |
95e344124 mm: remain migrat... |
927 |
set_freepage_migratetype(page, migratetype); |
c361be55b Move free pages b... |
928 |
page += 1 << order; |
d100313fd Fix calculation i... |
929 |
pages_moved += 1 << order; |
c361be55b Move free pages b... |
930 |
} |
d100313fd Fix calculation i... |
931 |
return pages_moved; |
c361be55b Move free pages b... |
932 |
} |
ee6f509c3 mm: factor out me... |
933 |
int move_freepages_block(struct zone *zone, struct page *page, |
68e3e9262 Revert "mm: compa... |
934 |
int migratetype) |
c361be55b Move free pages b... |
935 936 937 938 939 |
{ unsigned long start_pfn, end_pfn; struct page *start_page, *end_page; start_pfn = page_to_pfn(page); |
d9c234005 Do not depend on ... |
940 |
start_pfn = start_pfn & ~(pageblock_nr_pages-1); |
c361be55b Move free pages b... |
941 |
start_page = pfn_to_page(start_pfn); |
d9c234005 Do not depend on ... |
942 943 |
end_page = start_page + pageblock_nr_pages - 1; end_pfn = start_pfn + pageblock_nr_pages - 1; |
c361be55b Move free pages b... |
944 945 |
/* Do not cross zone boundaries */ |
108bcc96e mm: add & use zon... |
946 |
if (!zone_spans_pfn(zone, start_pfn)) |
c361be55b Move free pages b... |
947 |
start_page = page; |
108bcc96e mm: add & use zon... |
948 |
if (!zone_spans_pfn(zone, end_pfn)) |
c361be55b Move free pages b... |
949 950 951 952 |
return 0; return move_freepages(zone, start_page, end_page, migratetype); } |
2f66a68f3 page-allocator: c... |
953 954 955 956 957 958 959 960 961 962 |
static void change_pageblock_range(struct page *pageblock_page, int start_order, int migratetype) { int nr_pageblocks = 1 << (start_order - pageblock_order); while (nr_pageblocks--) { set_pageblock_migratetype(pageblock_page, migratetype); pageblock_page += pageblock_nr_pages; } } |
b2a0ac887 Split the free li... |
963 |
/* Remove an element from the buddy allocator from the fallback list */ |
0ac3a4099 page allocator: i... |
964 965 |
static inline struct page * __rmqueue_fallback(struct zone *zone, int order, int start_migratetype) |
b2a0ac887 Split the free li... |
966 967 968 969 970 971 972 973 974 |
{ struct free_area * area; int current_order; struct page *page; int migratetype, i; /* Find the largest possible block of pages in the other list */ for (current_order = MAX_ORDER-1; current_order >= order; --current_order) { |
6d4a49160 mm: page_alloc: c... |
975 |
for (i = 0;; i++) { |
b2a0ac887 Split the free li... |
976 |
migratetype = fallbacks[start_migratetype][i]; |
56fd56b86 Bias the location... |
977 978 |
/* MIGRATE_RESERVE handled later if necessary */ if (migratetype == MIGRATE_RESERVE) |
6d4a49160 mm: page_alloc: c... |
979 |
break; |
e010487db Group high-order ... |
980 |
|
b2a0ac887 Split the free li... |
981 982 983 984 985 986 987 988 989 |
area = &(zone->free_area[current_order]); if (list_empty(&area->free_list[migratetype])) continue; page = list_entry(area->free_list[migratetype].next, struct page, lru); area->nr_free--; /* |
c361be55b Move free pages b... |
990 |
* If breaking a large block of pages, move all free |
46dafbca2 Be more agressive... |
991 992 |
* pages to the preferred allocation list. If falling * back for a reclaimable kernel allocation, be more |
25985edce Fix common misspe... |
993 |
* aggressive about taking ownership of free pages |
47118af07 mm: mmzone: MIGRA... |
994 995 996 997 998 999 |
* * On the other hand, never change migration * type of MIGRATE_CMA pageblocks nor move CMA * pages on different free lists. We don't * want unmovable pages to be allocated from * MIGRATE_CMA areas. |
b2a0ac887 Split the free li... |
1000 |
*/ |
47118af07 mm: mmzone: MIGRA... |
1001 1002 1003 1004 1005 |
if (!is_migrate_cma(migratetype) && (unlikely(current_order >= pageblock_order / 2) || start_migratetype == MIGRATE_RECLAIMABLE || page_group_by_mobility_disabled)) { int pages; |
46dafbca2 Be more agressive... |
1006 1007 1008 1009 |
pages = move_freepages_block(zone, page, start_migratetype); /* Claim the whole block if over half of it is free */ |
dd5d241ea page-allocator: a... |
1010 1011 |
if (pages >= (1 << (pageblock_order-1)) || page_group_by_mobility_disabled) |
46dafbca2 Be more agressive... |
1012 1013 |
set_pageblock_migratetype(page, start_migratetype); |
b2a0ac887 Split the free li... |
1014 |
migratetype = start_migratetype; |
c361be55b Move free pages b... |
1015 |
} |
b2a0ac887 Split the free li... |
1016 1017 1018 1019 |
/* Remove the page from the freelists */ list_del(&page->lru); rmv_page_order(page); |
b2a0ac887 Split the free li... |
1020 |
|
2f66a68f3 page-allocator: c... |
1021 |
/* Take ownership for orders >= pageblock_order */ |
47118af07 mm: mmzone: MIGRA... |
1022 1023 |
if (current_order >= pageblock_order && !is_migrate_cma(migratetype)) |
2f66a68f3 page-allocator: c... |
1024 |
change_pageblock_range(page, current_order, |
b2a0ac887 Split the free li... |
1025 |
start_migratetype); |
47118af07 mm: mmzone: MIGRA... |
1026 1027 1028 |
expand(zone, page, order, current_order, area, is_migrate_cma(migratetype) ? migratetype : start_migratetype); |
e0fff1bd1 tracing, page-all... |
1029 1030 1031 |
trace_mm_page_alloc_extfrag(page, order, current_order, start_migratetype, migratetype); |
b2a0ac887 Split the free li... |
1032 1033 1034 |
return page; } } |
728ec980f page allocator: i... |
1035 |
return NULL; |
b2a0ac887 Split the free li... |
1036 |
} |
56fd56b86 Bias the location... |
1037 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1038 1039 1040 |
* Do the hard work of removing an element from the buddy allocator. * Call me with the zone->lock already held. */ |
b2a0ac887 Split the free li... |
1041 1042 |
static struct page *__rmqueue(struct zone *zone, unsigned int order, int migratetype) |
1da177e4c Linux-2.6.12-rc2 |
1043 |
{ |
1da177e4c Linux-2.6.12-rc2 |
1044 |
struct page *page; |
728ec980f page allocator: i... |
1045 |
retry_reserve: |
56fd56b86 Bias the location... |
1046 |
page = __rmqueue_smallest(zone, order, migratetype); |
b2a0ac887 Split the free li... |
1047 |
|
728ec980f page allocator: i... |
1048 |
if (unlikely(!page) && migratetype != MIGRATE_RESERVE) { |
56fd56b86 Bias the location... |
1049 |
page = __rmqueue_fallback(zone, order, migratetype); |
b2a0ac887 Split the free li... |
1050 |
|
728ec980f page allocator: i... |
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 |
/* * Use MIGRATE_RESERVE rather than fail an allocation. goto * is used because __rmqueue_smallest is an inline function * and we want just one call site */ if (!page) { migratetype = MIGRATE_RESERVE; goto retry_reserve; } } |
0d3d062a6 tracing, page-all... |
1061 |
trace_mm_page_alloc_zone_locked(page, order, migratetype); |
b2a0ac887 Split the free li... |
1062 |
return page; |
1da177e4c Linux-2.6.12-rc2 |
1063 |
} |
5f63b720b mm: page_alloc: r... |
1064 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1065 1066 1067 1068 |
* Obtain a specified number of elements from the buddy allocator, all under * a single hold of the lock, for efficiency. Add them to the supplied list. * Returns the number of new pages which were placed at *list. */ |
5f63b720b mm: page_alloc: r... |
1069 |
static int rmqueue_bulk(struct zone *zone, unsigned int order, |
b2a0ac887 Split the free li... |
1070 |
unsigned long count, struct list_head *list, |
e084b2d95 page-allocator: p... |
1071 |
int migratetype, int cold) |
1da177e4c Linux-2.6.12-rc2 |
1072 |
{ |
47118af07 mm: mmzone: MIGRA... |
1073 |
int mt = migratetype, i; |
5f63b720b mm: page_alloc: r... |
1074 |
|
c54ad30c7 [PATCH] mm: pagea... |
1075 |
spin_lock(&zone->lock); |
1da177e4c Linux-2.6.12-rc2 |
1076 |
for (i = 0; i < count; ++i) { |
b2a0ac887 Split the free li... |
1077 |
struct page *page = __rmqueue(zone, order, migratetype); |
085cc7d5d [PATCH] mm: page_... |
1078 |
if (unlikely(page == NULL)) |
1da177e4c Linux-2.6.12-rc2 |
1079 |
break; |
81eabcbe0 mm: fix page allo... |
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 |
/* * Split buddy pages returned by expand() are received here * in physical page order. The page is added to the callers and * list and the list head then moves forward. From the callers * perspective, the linked list is ordered by page number in * some conditions. This is useful for IO devices that can * merge IO requests if the physical pages are ordered * properly. */ |
e084b2d95 page-allocator: p... |
1090 1091 1092 1093 |
if (likely(cold == 0)) list_add(&page->lru, list); else list_add_tail(&page->lru, list); |
47118af07 mm: mmzone: MIGRA... |
1094 1095 |
if (IS_ENABLED(CONFIG_CMA)) { mt = get_pageblock_migratetype(page); |
194159fbc mm: remove MIGRAT... |
1096 |
if (!is_migrate_cma(mt) && !is_migrate_isolate(mt)) |
47118af07 mm: mmzone: MIGRA... |
1097 1098 |
mt = migratetype; } |
b12c4ad14 mm: page_alloc: u... |
1099 |
set_freepage_migratetype(page, mt); |
81eabcbe0 mm: fix page allo... |
1100 |
list = &page->lru; |
d1ce749a0 cma: count free C... |
1101 1102 1103 |
if (is_migrate_cma(mt)) __mod_zone_page_state(zone, NR_FREE_CMA_PAGES, -(1 << order)); |
1da177e4c Linux-2.6.12-rc2 |
1104 |
} |
f2260e6b1 page allocator: u... |
1105 |
__mod_zone_page_state(zone, NR_FREE_PAGES, -(i << order)); |
c54ad30c7 [PATCH] mm: pagea... |
1106 |
spin_unlock(&zone->lock); |
085cc7d5d [PATCH] mm: page_... |
1107 |
return i; |
1da177e4c Linux-2.6.12-rc2 |
1108 |
} |
4ae7c0394 [PATCH] Periodica... |
1109 |
#ifdef CONFIG_NUMA |
8fce4d8e3 [PATCH] slab: Nod... |
1110 |
/* |
4037d4522 Move remote node ... |
1111 1112 1113 1114 |
* Called from the vmstat counter updater to drain pagesets of this * currently executing processor on remote nodes after they have * expired. * |
879336c39 [PATCH] drain_nod... |
1115 1116 |
* Note that this function must be called with the thread pinned to * a single processor. |
8fce4d8e3 [PATCH] slab: Nod... |
1117 |
*/ |
4037d4522 Move remote node ... |
1118 |
void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp) |
4ae7c0394 [PATCH] Periodica... |
1119 |
{ |
4ae7c0394 [PATCH] Periodica... |
1120 |
unsigned long flags; |
4037d4522 Move remote node ... |
1121 |
int to_drain; |
4ae7c0394 [PATCH] Periodica... |
1122 |
|
4037d4522 Move remote node ... |
1123 1124 1125 1126 1127 |
local_irq_save(flags); if (pcp->count >= pcp->batch) to_drain = pcp->batch; else to_drain = pcp->count; |
2a13515c3 mm: clear pages_s... |
1128 1129 1130 1131 |
if (to_drain > 0) { free_pcppages_bulk(zone, to_drain, pcp); pcp->count -= to_drain; } |
4037d4522 Move remote node ... |
1132 |
local_irq_restore(flags); |
4ae7c0394 [PATCH] Periodica... |
1133 1134 |
} #endif |
9f8f21725 Page allocator: c... |
1135 1136 1137 1138 1139 1140 1141 1142 |
/* * Drain pages of the indicated processor. * * The processor must either be the current processor and the * thread pinned to the current processor or a processor that * is not online. */ static void drain_pages(unsigned int cpu) |
1da177e4c Linux-2.6.12-rc2 |
1143 |
{ |
c54ad30c7 [PATCH] mm: pagea... |
1144 |
unsigned long flags; |
1da177e4c Linux-2.6.12-rc2 |
1145 |
struct zone *zone; |
1da177e4c Linux-2.6.12-rc2 |
1146 |
|
ee99c71c5 mm: introduce for... |
1147 |
for_each_populated_zone(zone) { |
1da177e4c Linux-2.6.12-rc2 |
1148 |
struct per_cpu_pageset *pset; |
3dfa5721f Page allocator: g... |
1149 |
struct per_cpu_pages *pcp; |
1da177e4c Linux-2.6.12-rc2 |
1150 |
|
99dcc3e5a this_cpu: Page al... |
1151 1152 |
local_irq_save(flags); pset = per_cpu_ptr(zone->pageset, cpu); |
3dfa5721f Page allocator: g... |
1153 1154 |
pcp = &pset->pcp; |
2ff754fa8 mm: clear pages_s... |
1155 1156 1157 1158 |
if (pcp->count) { free_pcppages_bulk(zone, pcp->count, pcp); pcp->count = 0; } |
3dfa5721f Page allocator: g... |
1159 |
local_irq_restore(flags); |
1da177e4c Linux-2.6.12-rc2 |
1160 1161 |
} } |
1da177e4c Linux-2.6.12-rc2 |
1162 |
|
9f8f21725 Page allocator: c... |
1163 1164 1165 1166 1167 1168 1169 1170 1171 |
/* * Spill all of this CPU's per-cpu pages back into the buddy allocator. */ void drain_local_pages(void *arg) { drain_pages(smp_processor_id()); } /* |
74046494e mm: only IPI CPUs... |
1172 1173 1174 1175 1176 1177 1178 |
* Spill all the per-cpu pages from all CPUs back into the buddy allocator. * * Note that this code is protected against sending an IPI to an offline * CPU but does not guarantee sending an IPI to newly hotplugged CPUs: * on_each_cpu_mask() blocks hotplug and won't talk to offlined CPUs but * nothing keeps CPUs from showing up after we populated the cpumask and * before the call to on_each_cpu_mask(). |
9f8f21725 Page allocator: c... |
1179 1180 1181 |
*/ void drain_all_pages(void) { |
74046494e mm: only IPI CPUs... |
1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 |
int cpu; struct per_cpu_pageset *pcp; struct zone *zone; /* * Allocate in the BSS so we wont require allocation in * direct reclaim path for CONFIG_CPUMASK_OFFSTACK=y */ static cpumask_t cpus_with_pcps; /* * We don't care about racing with CPU hotplug event * as offline notification will cause the notified * cpu to drain that CPU pcps and on_each_cpu_mask * disables preemption as part of its processing */ for_each_online_cpu(cpu) { bool has_pcps = false; for_each_populated_zone(zone) { pcp = per_cpu_ptr(zone->pageset, cpu); if (pcp->pcp.count) { has_pcps = true; break; } } if (has_pcps) cpumask_set_cpu(cpu, &cpus_with_pcps); else cpumask_clear_cpu(cpu, &cpus_with_pcps); } on_each_cpu_mask(&cpus_with_pcps, drain_local_pages, NULL, 1); |
9f8f21725 Page allocator: c... |
1213 |
} |
296699de6 Introduce CONFIG_... |
1214 |
#ifdef CONFIG_HIBERNATION |
1da177e4c Linux-2.6.12-rc2 |
1215 1216 1217 |
void mark_free_pages(struct zone *zone) { |
f623f0db8 [PATCH] swsusp: F... |
1218 1219 |
unsigned long pfn, max_zone_pfn; unsigned long flags; |
b2a0ac887 Split the free li... |
1220 |
int order, t; |
1da177e4c Linux-2.6.12-rc2 |
1221 1222 1223 1224 1225 1226 |
struct list_head *curr; if (!zone->spanned_pages) return; spin_lock_irqsave(&zone->lock, flags); |
f623f0db8 [PATCH] swsusp: F... |
1227 |
|
108bcc96e mm: add & use zon... |
1228 |
max_zone_pfn = zone_end_pfn(zone); |
f623f0db8 [PATCH] swsusp: F... |
1229 1230 1231 |
for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) if (pfn_valid(pfn)) { struct page *page = pfn_to_page(pfn); |
7be982349 swsusp: use inlin... |
1232 1233 |
if (!swsusp_page_is_forbidden(page)) swsusp_unset_page_free(page); |
f623f0db8 [PATCH] swsusp: F... |
1234 |
} |
1da177e4c Linux-2.6.12-rc2 |
1235 |
|
b2a0ac887 Split the free li... |
1236 1237 |
for_each_migratetype_order(order, t) { list_for_each(curr, &zone->free_area[order].free_list[t]) { |
f623f0db8 [PATCH] swsusp: F... |
1238 |
unsigned long i; |
1da177e4c Linux-2.6.12-rc2 |
1239 |
|
f623f0db8 [PATCH] swsusp: F... |
1240 1241 |
pfn = page_to_pfn(list_entry(curr, struct page, lru)); for (i = 0; i < (1UL << order); i++) |
7be982349 swsusp: use inlin... |
1242 |
swsusp_set_page_free(pfn_to_page(pfn + i)); |
f623f0db8 [PATCH] swsusp: F... |
1243 |
} |
b2a0ac887 Split the free li... |
1244 |
} |
1da177e4c Linux-2.6.12-rc2 |
1245 1246 |
spin_unlock_irqrestore(&zone->lock, flags); } |
e2c55dc87 Drain per-cpu lis... |
1247 |
#endif /* CONFIG_PM */ |
1da177e4c Linux-2.6.12-rc2 |
1248 1249 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1250 |
* Free a 0-order page |
fc91668ea mm: remove free_h... |
1251 |
* cold == 1 ? free a cold page : free a hot page |
1da177e4c Linux-2.6.12-rc2 |
1252 |
*/ |
fc91668ea mm: remove free_h... |
1253 |
void free_hot_cold_page(struct page *page, int cold) |
1da177e4c Linux-2.6.12-rc2 |
1254 1255 1256 1257 |
{ struct zone *zone = page_zone(page); struct per_cpu_pages *pcp; unsigned long flags; |
5f8dcc212 page-allocator: s... |
1258 |
int migratetype; |
1da177e4c Linux-2.6.12-rc2 |
1259 |
|
ec95f53aa mm: introduce fre... |
1260 |
if (!free_pages_prepare(page, 0)) |
689bcebfd [PATCH] unpaged: ... |
1261 |
return; |
5f8dcc212 page-allocator: s... |
1262 |
migratetype = get_pageblock_migratetype(page); |
b12c4ad14 mm: page_alloc: u... |
1263 |
set_freepage_migratetype(page, migratetype); |
1da177e4c Linux-2.6.12-rc2 |
1264 |
local_irq_save(flags); |
f8891e5e1 [PATCH] Light wei... |
1265 |
__count_vm_event(PGFREE); |
da456f14d page allocator: d... |
1266 |
|
5f8dcc212 page-allocator: s... |
1267 1268 1269 1270 1271 1272 1273 1274 |
/* * We only track unmovable, reclaimable and movable on pcp lists. * Free ISOLATE pages back to the allocator because they are being * offlined but treat RESERVE as movable pages so we can get those * areas back if necessary. Otherwise, we may have to free * excessively into the page allocator */ if (migratetype >= MIGRATE_PCPTYPES) { |
194159fbc mm: remove MIGRAT... |
1275 |
if (unlikely(is_migrate_isolate(migratetype))) { |
5f8dcc212 page-allocator: s... |
1276 1277 1278 1279 1280 |
free_one_page(zone, page, 0, migratetype); goto out; } migratetype = MIGRATE_MOVABLE; } |
99dcc3e5a this_cpu: Page al... |
1281 |
pcp = &this_cpu_ptr(zone->pageset)->pcp; |
3dfa5721f Page allocator: g... |
1282 |
if (cold) |
5f8dcc212 page-allocator: s... |
1283 |
list_add_tail(&page->lru, &pcp->lists[migratetype]); |
3dfa5721f Page allocator: g... |
1284 |
else |
5f8dcc212 page-allocator: s... |
1285 |
list_add(&page->lru, &pcp->lists[migratetype]); |
1da177e4c Linux-2.6.12-rc2 |
1286 |
pcp->count++; |
48db57f8f [PATCH] mm: free_... |
1287 |
if (pcp->count >= pcp->high) { |
5f8dcc212 page-allocator: s... |
1288 |
free_pcppages_bulk(zone, pcp->batch, pcp); |
48db57f8f [PATCH] mm: free_... |
1289 1290 |
pcp->count -= pcp->batch; } |
5f8dcc212 page-allocator: s... |
1291 1292 |
out: |
1da177e4c Linux-2.6.12-rc2 |
1293 |
local_irq_restore(flags); |
1da177e4c Linux-2.6.12-rc2 |
1294 |
} |
8dfcc9ba2 [PATCH] mm: split... |
1295 |
/* |
cc59850ef mm: add free_hot_... |
1296 1297 1298 1299 1300 1301 1302 |
* Free a list of 0-order pages */ void free_hot_cold_page_list(struct list_head *list, int cold) { struct page *page, *next; list_for_each_entry_safe(page, next, list, lru) { |
b413d48aa mm-tracepoint: re... |
1303 |
trace_mm_page_free_batched(page, cold); |
cc59850ef mm: add free_hot_... |
1304 1305 1306 1307 1308 |
free_hot_cold_page(page, cold); } } /* |
8dfcc9ba2 [PATCH] mm: split... |
1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 |
* split_page takes a non-compound higher-order page, and splits it into * n (1<<order) sub-pages: page[0..n] * Each sub-page must be freed individually. * * Note: this is probably too low level an operation for use in drivers. * Please consult with lkml before using this in your driver. */ void split_page(struct page *page, unsigned int order) { int i; |
725d704ec [PATCH] mm: VM_BU... |
1319 1320 |
VM_BUG_ON(PageCompound(page)); VM_BUG_ON(!page_count(page)); |
b1eeab676 kmemcheck: add ho... |
1321 1322 1323 1324 1325 1326 1327 1328 1329 |
#ifdef CONFIG_KMEMCHECK /* * Split shadow pages too, because free(page[0]) would * otherwise free the whole shadow. */ if (kmemcheck_page_is_tracked(page)) split_page(virt_to_page(page[0].shadow), order); #endif |
7835e98b2 [PATCH] remove se... |
1330 1331 |
for (i = 1; i < (1 << order); i++) set_page_refcounted(page + i); |
8dfcc9ba2 [PATCH] mm: split... |
1332 |
} |
5853ff23c mm: export split_... |
1333 |
EXPORT_SYMBOL_GPL(split_page); |
8dfcc9ba2 [PATCH] mm: split... |
1334 |
|
8fb74b9fb mm: compaction: p... |
1335 |
static int __isolate_free_page(struct page *page, unsigned int order) |
748446bb6 mm: compaction: m... |
1336 |
{ |
748446bb6 mm: compaction: m... |
1337 1338 |
unsigned long watermark; struct zone *zone; |
2139cbe62 cma: fix counting... |
1339 |
int mt; |
748446bb6 mm: compaction: m... |
1340 1341 1342 1343 |
BUG_ON(!PageBuddy(page)); zone = page_zone(page); |
2e30abd17 mm: cma: skip wat... |
1344 |
mt = get_pageblock_migratetype(page); |
748446bb6 mm: compaction: m... |
1345 |
|
194159fbc mm: remove MIGRAT... |
1346 |
if (!is_migrate_isolate(mt)) { |
2e30abd17 mm: cma: skip wat... |
1347 1348 1349 1350 |
/* Obey watermarks as if the page was being allocated */ watermark = low_wmark_pages(zone) + (1 << order); if (!zone_watermark_ok(zone, 0, watermark, 0, 0)) return 0; |
8fb74b9fb mm: compaction: p... |
1351 |
__mod_zone_freepage_state(zone, -(1UL << order), mt); |
2e30abd17 mm: cma: skip wat... |
1352 |
} |
748446bb6 mm: compaction: m... |
1353 1354 1355 1356 1357 |
/* Remove page from free list */ list_del(&page->lru); zone->free_area[order].nr_free--; rmv_page_order(page); |
2139cbe62 cma: fix counting... |
1358 |
|
8fb74b9fb mm: compaction: p... |
1359 |
/* Set the pageblock if the isolated page is at least a pageblock */ |
748446bb6 mm: compaction: m... |
1360 1361 |
if (order >= pageblock_order - 1) { struct page *endpage = page + (1 << order) - 1; |
47118af07 mm: mmzone: MIGRA... |
1362 1363 |
for (; page < endpage; page += pageblock_nr_pages) { int mt = get_pageblock_migratetype(page); |
194159fbc mm: remove MIGRAT... |
1364 |
if (!is_migrate_isolate(mt) && !is_migrate_cma(mt)) |
47118af07 mm: mmzone: MIGRA... |
1365 1366 1367 |
set_pageblock_migratetype(page, MIGRATE_MOVABLE); } |
748446bb6 mm: compaction: m... |
1368 |
} |
8fb74b9fb mm: compaction: p... |
1369 |
return 1UL << order; |
1fb3f8ca0 mm: compaction: c... |
1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 |
} /* * Similar to split_page except the page is already free. As this is only * being used for migration, the migratetype of the block also changes. * As this is called with interrupts disabled, the caller is responsible * for calling arch_alloc_page() and kernel_map_page() after interrupts * are enabled. * * Note: this is probably too low level an operation for use in drivers. * Please consult with lkml before using this in your driver. */ int split_free_page(struct page *page) { unsigned int order; int nr_pages; |
1fb3f8ca0 mm: compaction: c... |
1386 |
order = page_order(page); |
8fb74b9fb mm: compaction: p... |
1387 |
nr_pages = __isolate_free_page(page, order); |
1fb3f8ca0 mm: compaction: c... |
1388 1389 1390 1391 1392 1393 1394 |
if (!nr_pages) return 0; /* Split into individual pages */ set_page_refcounted(page); split_page(page, order); return nr_pages; |
748446bb6 mm: compaction: m... |
1395 1396 1397 |
} /* |
1da177e4c Linux-2.6.12-rc2 |
1398 1399 1400 1401 |
* Really, prep_compound_page() should be called from __rmqueue_bulk(). But * we cheat by calling it from here, in the order > 0 path. Saves a branch * or two. */ |
0a15c3e9f page allocator: i... |
1402 1403 |
static inline struct page *buffered_rmqueue(struct zone *preferred_zone, |
3dd282669 page allocator: c... |
1404 1405 |
struct zone *zone, int order, gfp_t gfp_flags, int migratetype) |
1da177e4c Linux-2.6.12-rc2 |
1406 1407 |
{ unsigned long flags; |
689bcebfd [PATCH] unpaged: ... |
1408 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
1409 |
int cold = !!(gfp_flags & __GFP_COLD); |
689bcebfd [PATCH] unpaged: ... |
1410 |
again: |
48db57f8f [PATCH] mm: free_... |
1411 |
if (likely(order == 0)) { |
1da177e4c Linux-2.6.12-rc2 |
1412 |
struct per_cpu_pages *pcp; |
5f8dcc212 page-allocator: s... |
1413 |
struct list_head *list; |
1da177e4c Linux-2.6.12-rc2 |
1414 |
|
1da177e4c Linux-2.6.12-rc2 |
1415 |
local_irq_save(flags); |
99dcc3e5a this_cpu: Page al... |
1416 1417 |
pcp = &this_cpu_ptr(zone->pageset)->pcp; list = &pcp->lists[migratetype]; |
5f8dcc212 page-allocator: s... |
1418 |
if (list_empty(list)) { |
535131e69 Choose pages from... |
1419 |
pcp->count += rmqueue_bulk(zone, 0, |
5f8dcc212 page-allocator: s... |
1420 |
pcp->batch, list, |
e084b2d95 page-allocator: p... |
1421 |
migratetype, cold); |
5f8dcc212 page-allocator: s... |
1422 |
if (unlikely(list_empty(list))) |
6fb332fab memory hotplug: e... |
1423 |
goto failed; |
535131e69 Choose pages from... |
1424 |
} |
b92a6edd4 Add a configure o... |
1425 |
|
5f8dcc212 page-allocator: s... |
1426 1427 1428 1429 |
if (cold) page = list_entry(list->prev, struct page, lru); else page = list_entry(list->next, struct page, lru); |
b92a6edd4 Add a configure o... |
1430 1431 |
list_del(&page->lru); pcp->count--; |
7fb1d9fca [PATCH] mm: __all... |
1432 |
} else { |
dab48dab3 page-allocator: w... |
1433 1434 1435 1436 1437 1438 1439 1440 |
if (unlikely(gfp_flags & __GFP_NOFAIL)) { /* * __GFP_NOFAIL is not to be used in new code. * * All __GFP_NOFAIL callers should be fixed so that they * properly detect and handle allocation failures. * * We most definitely don't want callers attempting to |
4923abf9f Don't warn about ... |
1441 |
* allocate greater than order-1 page units with |
dab48dab3 page-allocator: w... |
1442 1443 |
* __GFP_NOFAIL. */ |
4923abf9f Don't warn about ... |
1444 |
WARN_ON_ONCE(order > 1); |
dab48dab3 page-allocator: w... |
1445 |
} |
1da177e4c Linux-2.6.12-rc2 |
1446 |
spin_lock_irqsave(&zone->lock, flags); |
b2a0ac887 Split the free li... |
1447 |
page = __rmqueue(zone, order, migratetype); |
a74609faf [PATCH] mm: page_... |
1448 1449 1450 |
spin_unlock(&zone->lock); if (!page) goto failed; |
d1ce749a0 cma: count free C... |
1451 1452 |
__mod_zone_freepage_state(zone, -(1 << order), get_pageblock_migratetype(page)); |
1da177e4c Linux-2.6.12-rc2 |
1453 |
} |
f8891e5e1 [PATCH] Light wei... |
1454 |
__count_zone_vm_events(PGALLOC, zone, 1 << order); |
78afd5612 mm: add __GFP_OTH... |
1455 |
zone_statistics(preferred_zone, zone, gfp_flags); |
a74609faf [PATCH] mm: page_... |
1456 |
local_irq_restore(flags); |
1da177e4c Linux-2.6.12-rc2 |
1457 |
|
725d704ec [PATCH] mm: VM_BU... |
1458 |
VM_BUG_ON(bad_range(zone, page)); |
17cf44064 [PATCH] mm: clean... |
1459 |
if (prep_new_page(page, order, gfp_flags)) |
a74609faf [PATCH] mm: page_... |
1460 |
goto again; |
1da177e4c Linux-2.6.12-rc2 |
1461 |
return page; |
a74609faf [PATCH] mm: page_... |
1462 1463 1464 |
failed: local_irq_restore(flags); |
a74609faf [PATCH] mm: page_... |
1465 |
return NULL; |
1da177e4c Linux-2.6.12-rc2 |
1466 |
} |
933e312e7 [PATCH] fault-inj... |
1467 |
#ifdef CONFIG_FAIL_PAGE_ALLOC |
b2588c4b4 fail_page_alloc: ... |
1468 |
static struct { |
933e312e7 [PATCH] fault-inj... |
1469 1470 1471 1472 |
struct fault_attr attr; u32 ignore_gfp_highmem; u32 ignore_gfp_wait; |
54114994f fault-injection: ... |
1473 |
u32 min_order; |
933e312e7 [PATCH] fault-inj... |
1474 1475 |
} fail_page_alloc = { .attr = FAULT_ATTR_INITIALIZER, |
6b1b60f41 [PATCH] fault-inj... |
1476 1477 |
.ignore_gfp_wait = 1, .ignore_gfp_highmem = 1, |
54114994f fault-injection: ... |
1478 |
.min_order = 1, |
933e312e7 [PATCH] fault-inj... |
1479 1480 1481 1482 1483 1484 1485 |
}; static int __init setup_fail_page_alloc(char *str) { return setup_fault_attr(&fail_page_alloc.attr, str); } __setup("fail_page_alloc=", setup_fail_page_alloc); |
deaf386ee mm/buddy: cleanup... |
1486 |
static bool should_fail_alloc_page(gfp_t gfp_mask, unsigned int order) |
933e312e7 [PATCH] fault-inj... |
1487 |
{ |
54114994f fault-injection: ... |
1488 |
if (order < fail_page_alloc.min_order) |
deaf386ee mm/buddy: cleanup... |
1489 |
return false; |
933e312e7 [PATCH] fault-inj... |
1490 |
if (gfp_mask & __GFP_NOFAIL) |
deaf386ee mm/buddy: cleanup... |
1491 |
return false; |
933e312e7 [PATCH] fault-inj... |
1492 |
if (fail_page_alloc.ignore_gfp_highmem && (gfp_mask & __GFP_HIGHMEM)) |
deaf386ee mm/buddy: cleanup... |
1493 |
return false; |
933e312e7 [PATCH] fault-inj... |
1494 |
if (fail_page_alloc.ignore_gfp_wait && (gfp_mask & __GFP_WAIT)) |
deaf386ee mm/buddy: cleanup... |
1495 |
return false; |
933e312e7 [PATCH] fault-inj... |
1496 1497 1498 1499 1500 1501 1502 1503 |
return should_fail(&fail_page_alloc.attr, 1 << order); } #ifdef CONFIG_FAULT_INJECTION_DEBUG_FS static int __init fail_page_alloc_debugfs(void) { |
f4ae40a6a switch debugfs to... |
1504 |
umode_t mode = S_IFREG | S_IRUSR | S_IWUSR; |
933e312e7 [PATCH] fault-inj... |
1505 |
struct dentry *dir; |
933e312e7 [PATCH] fault-inj... |
1506 |
|
dd48c085c fault-injection: ... |
1507 1508 1509 1510 |
dir = fault_create_debugfs_attr("fail_page_alloc", NULL, &fail_page_alloc.attr); if (IS_ERR(dir)) return PTR_ERR(dir); |
933e312e7 [PATCH] fault-inj... |
1511 |
|
b2588c4b4 fail_page_alloc: ... |
1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 |
if (!debugfs_create_bool("ignore-gfp-wait", mode, dir, &fail_page_alloc.ignore_gfp_wait)) goto fail; if (!debugfs_create_bool("ignore-gfp-highmem", mode, dir, &fail_page_alloc.ignore_gfp_highmem)) goto fail; if (!debugfs_create_u32("min-order", mode, dir, &fail_page_alloc.min_order)) goto fail; return 0; fail: |
dd48c085c fault-injection: ... |
1524 |
debugfs_remove_recursive(dir); |
933e312e7 [PATCH] fault-inj... |
1525 |
|
b2588c4b4 fail_page_alloc: ... |
1526 |
return -ENOMEM; |
933e312e7 [PATCH] fault-inj... |
1527 1528 1529 1530 1531 1532 1533 |
} late_initcall(fail_page_alloc_debugfs); #endif /* CONFIG_FAULT_INJECTION_DEBUG_FS */ #else /* CONFIG_FAIL_PAGE_ALLOC */ |
deaf386ee mm/buddy: cleanup... |
1534 |
static inline bool should_fail_alloc_page(gfp_t gfp_mask, unsigned int order) |
933e312e7 [PATCH] fault-inj... |
1535 |
{ |
deaf386ee mm/buddy: cleanup... |
1536 |
return false; |
933e312e7 [PATCH] fault-inj... |
1537 1538 1539 |
} #endif /* CONFIG_FAIL_PAGE_ALLOC */ |
1da177e4c Linux-2.6.12-rc2 |
1540 |
/* |
88f5acf88 mm: page allocato... |
1541 |
* Return true if free pages are above 'mark'. This takes into account the order |
1da177e4c Linux-2.6.12-rc2 |
1542 1543 |
* of the allocation. */ |
88f5acf88 mm: page allocato... |
1544 1545 |
static bool __zone_watermark_ok(struct zone *z, int order, unsigned long mark, int classzone_idx, int alloc_flags, long free_pages) |
1da177e4c Linux-2.6.12-rc2 |
1546 1547 |
{ /* free_pages my go negative - that's OK */ |
d23ad4232 [PATCH] Use ZVC f... |
1548 |
long min = mark; |
2cfed0752 mm: fix free page... |
1549 |
long lowmem_reserve = z->lowmem_reserve[classzone_idx]; |
1da177e4c Linux-2.6.12-rc2 |
1550 |
int o; |
026b08147 mm/page_alloc.c: ... |
1551 |
long free_cma = 0; |
1da177e4c Linux-2.6.12-rc2 |
1552 |
|
df0a6daa0 mm: fix off-by-tw... |
1553 |
free_pages -= (1 << order) - 1; |
7fb1d9fca [PATCH] mm: __all... |
1554 |
if (alloc_flags & ALLOC_HIGH) |
1da177e4c Linux-2.6.12-rc2 |
1555 |
min -= min / 2; |
7fb1d9fca [PATCH] mm: __all... |
1556 |
if (alloc_flags & ALLOC_HARDER) |
1da177e4c Linux-2.6.12-rc2 |
1557 |
min -= min / 4; |
d95ea5d18 cma: fix watermar... |
1558 1559 1560 |
#ifdef CONFIG_CMA /* If allocation can't use CMA areas don't use free CMA pages */ if (!(alloc_flags & ALLOC_CMA)) |
026b08147 mm/page_alloc.c: ... |
1561 |
free_cma = zone_page_state(z, NR_FREE_CMA_PAGES); |
d95ea5d18 cma: fix watermar... |
1562 |
#endif |
026b08147 mm/page_alloc.c: ... |
1563 1564 |
if (free_pages - free_cma <= min + lowmem_reserve) |
88f5acf88 mm: page allocato... |
1565 |
return false; |
1da177e4c Linux-2.6.12-rc2 |
1566 1567 1568 1569 1570 1571 1572 1573 |
for (o = 0; o < order; o++) { /* At the next order, this order's pages become unavailable */ free_pages -= z->free_area[o].nr_free << o; /* Require fewer higher order pages to be free */ min >>= 1; if (free_pages <= min) |
88f5acf88 mm: page allocato... |
1574 |
return false; |
1da177e4c Linux-2.6.12-rc2 |
1575 |
} |
88f5acf88 mm: page allocato... |
1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 |
return true; } bool zone_watermark_ok(struct zone *z, int order, unsigned long mark, int classzone_idx, int alloc_flags) { return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags, zone_page_state(z, NR_FREE_PAGES)); } bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark, int classzone_idx, int alloc_flags) { long free_pages = zone_page_state(z, NR_FREE_PAGES); if (z->percpu_drift_mark && free_pages < z->percpu_drift_mark) free_pages = zone_page_state_snapshot(z, NR_FREE_PAGES); return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags, free_pages); |
1da177e4c Linux-2.6.12-rc2 |
1596 |
} |
9276b1bc9 [PATCH] memory pa... |
1597 1598 1599 1600 1601 1602 |
#ifdef CONFIG_NUMA /* * zlc_setup - Setup for "zonelist cache". Uses cached zone data to * skip over zones that are not allowed by the cpuset, or that have * been recently (in last second) found to be nearly full. See further * comments in mmzone.h. Reduces cache footprint of zonelist scans |
183ff22bb spelling fixes: mm/ |
1603 |
* that have to skip over a lot of full or unallowed zones. |
9276b1bc9 [PATCH] memory pa... |
1604 1605 1606 |
* * If the zonelist cache is present in the passed in zonelist, then * returns a pointer to the allowed node mask (either the current |
4b0ef1fe8 page_alloc: use N... |
1607 |
* tasks mems_allowed, or node_states[N_MEMORY].) |
9276b1bc9 [PATCH] memory pa... |
1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 |
* * If the zonelist cache is not available for this zonelist, does * nothing and returns NULL. * * If the fullzones BITMAP in the zonelist cache is stale (more than * a second since last zap'd) then we zap it out (clear its bits.) * * We hold off even calling zlc_setup, until after we've checked the * first zone in the zonelist, on the theory that most allocations will * be satisfied from that first zone, so best to examine that zone as * quickly as we can. */ static nodemask_t *zlc_setup(struct zonelist *zonelist, int alloc_flags) { struct zonelist_cache *zlc; /* cached zonelist speedup info */ nodemask_t *allowednodes; /* zonelist_cache approximation */ zlc = zonelist->zlcache_ptr; if (!zlc) return NULL; |
f05111f50 mm/page_alloc.c: ... |
1628 |
if (time_after(jiffies, zlc->last_full_zap + HZ)) { |
9276b1bc9 [PATCH] memory pa... |
1629 1630 1631 1632 1633 1634 |
bitmap_zero(zlc->fullzones, MAX_ZONES_PER_ZONELIST); zlc->last_full_zap = jiffies; } allowednodes = !in_interrupt() && (alloc_flags & ALLOC_CPUSET) ? &cpuset_current_mems_allowed : |
4b0ef1fe8 page_alloc: use N... |
1635 |
&node_states[N_MEMORY]; |
9276b1bc9 [PATCH] memory pa... |
1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 |
return allowednodes; } /* * Given 'z' scanning a zonelist, run a couple of quick checks to see * if it is worth looking at further for free memory: * 1) Check that the zone isn't thought to be full (doesn't have its * bit set in the zonelist_cache fullzones BITMAP). * 2) Check that the zones node (obtained from the zonelist_cache * z_to_n[] mapping) is allowed in the passed in allowednodes mask. * Return true (non-zero) if zone is worth looking at further, or * else return false (zero) if it is not. * * This check -ignores- the distinction between various watermarks, * such as GFP_HIGH, GFP_ATOMIC, PF_MEMALLOC, ... If a zone is * found to be full for any variation of these watermarks, it will * be considered full for up to one second by all requests, unless * we are so low on memory on all allowed nodes that we are forced * into the second scan of the zonelist. * * In the second scan we ignore this zonelist cache and exactly * apply the watermarks to all zones, even it is slower to do so. * We are low on memory in the second scan, and should leave no stone * unturned looking for a free page. */ |
dd1a239f6 mm: have zonelist... |
1661 |
static int zlc_zone_worth_trying(struct zonelist *zonelist, struct zoneref *z, |
9276b1bc9 [PATCH] memory pa... |
1662 1663 1664 1665 1666 1667 1668 1669 1670 |
nodemask_t *allowednodes) { struct zonelist_cache *zlc; /* cached zonelist speedup info */ int i; /* index of *z in zonelist zones */ int n; /* node that zone *z is on */ zlc = zonelist->zlcache_ptr; if (!zlc) return 1; |
dd1a239f6 mm: have zonelist... |
1671 |
i = z - zonelist->_zonerefs; |
9276b1bc9 [PATCH] memory pa... |
1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 |
n = zlc->z_to_n[i]; /* This zone is worth trying if it is allowed but not full */ return node_isset(n, *allowednodes) && !test_bit(i, zlc->fullzones); } /* * Given 'z' scanning a zonelist, set the corresponding bit in * zlc->fullzones, so that subsequent attempts to allocate a page * from that zone don't waste time re-examining it. */ |
dd1a239f6 mm: have zonelist... |
1683 |
static void zlc_mark_zone_full(struct zonelist *zonelist, struct zoneref *z) |
9276b1bc9 [PATCH] memory pa... |
1684 1685 1686 1687 1688 1689 1690 |
{ struct zonelist_cache *zlc; /* cached zonelist speedup info */ int i; /* index of *z in zonelist zones */ zlc = zonelist->zlcache_ptr; if (!zlc) return; |
dd1a239f6 mm: have zonelist... |
1691 |
i = z - zonelist->_zonerefs; |
9276b1bc9 [PATCH] memory pa... |
1692 1693 1694 |
set_bit(i, zlc->fullzones); } |
76d3fbf8f mm: page allocato... |
1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 |
/* * clear all zones full, called after direct reclaim makes progress so that * a zone that was recently full is not skipped over for up to a second */ static void zlc_clear_zones_full(struct zonelist *zonelist) { struct zonelist_cache *zlc; /* cached zonelist speedup info */ zlc = zonelist->zlcache_ptr; if (!zlc) return; bitmap_zero(zlc->fullzones, MAX_ZONES_PER_ZONELIST); } |
957f822a0 mm, numa: reclaim... |
1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 |
static bool zone_allows_reclaim(struct zone *local_zone, struct zone *zone) { return node_isset(local_zone->node, zone->zone_pgdat->reclaim_nodes); } static void __paginginit init_zone_allows_reclaim(int nid) { int i; for_each_online_node(i) |
6b187d026 mm, numa: avoid s... |
1719 |
if (node_distance(nid, i) <= RECLAIM_DISTANCE) |
957f822a0 mm, numa: reclaim... |
1720 |
node_set(i, NODE_DATA(nid)->reclaim_nodes); |
6b187d026 mm, numa: avoid s... |
1721 |
else |
957f822a0 mm, numa: reclaim... |
1722 |
zone_reclaim_mode = 1; |
957f822a0 mm, numa: reclaim... |
1723 |
} |
9276b1bc9 [PATCH] memory pa... |
1724 1725 1726 1727 1728 1729 |
#else /* CONFIG_NUMA */ static nodemask_t *zlc_setup(struct zonelist *zonelist, int alloc_flags) { return NULL; } |
dd1a239f6 mm: have zonelist... |
1730 |
static int zlc_zone_worth_trying(struct zonelist *zonelist, struct zoneref *z, |
9276b1bc9 [PATCH] memory pa... |
1731 1732 1733 1734 |
nodemask_t *allowednodes) { return 1; } |
dd1a239f6 mm: have zonelist... |
1735 |
static void zlc_mark_zone_full(struct zonelist *zonelist, struct zoneref *z) |
9276b1bc9 [PATCH] memory pa... |
1736 1737 |
{ } |
76d3fbf8f mm: page allocato... |
1738 1739 1740 1741 |
static void zlc_clear_zones_full(struct zonelist *zonelist) { } |
957f822a0 mm, numa: reclaim... |
1742 1743 1744 1745 1746 1747 1748 1749 1750 |
static bool zone_allows_reclaim(struct zone *local_zone, struct zone *zone) { return true; } static inline void init_zone_allows_reclaim(int nid) { } |
9276b1bc9 [PATCH] memory pa... |
1751 |
#endif /* CONFIG_NUMA */ |
7fb1d9fca [PATCH] mm: __all... |
1752 |
/* |
0798e5193 [PATCH] memory pa... |
1753 |
* get_page_from_freelist goes through the zonelist trying to allocate |
7fb1d9fca [PATCH] mm: __all... |
1754 1755 1756 |
* a page. */ static struct page * |
19770b326 mm: filter based ... |
1757 |
get_page_from_freelist(gfp_t gfp_mask, nodemask_t *nodemask, unsigned int order, |
5117f45d1 page allocator: c... |
1758 |
struct zonelist *zonelist, int high_zoneidx, int alloc_flags, |
3dd282669 page allocator: c... |
1759 |
struct zone *preferred_zone, int migratetype) |
753ee7289 [PATCH] VM: early... |
1760 |
{ |
dd1a239f6 mm: have zonelist... |
1761 |
struct zoneref *z; |
7fb1d9fca [PATCH] mm: __all... |
1762 |
struct page *page = NULL; |
54a6eb5c4 mm: use two zonel... |
1763 |
int classzone_idx; |
5117f45d1 page allocator: c... |
1764 |
struct zone *zone; |
9276b1bc9 [PATCH] memory pa... |
1765 1766 1767 |
nodemask_t *allowednodes = NULL;/* zonelist_cache approximation */ int zlc_active = 0; /* set if using zonelist_cache */ int did_zlc_setup = 0; /* just call zlc_setup() one time */ |
54a6eb5c4 mm: use two zonel... |
1768 |
|
19770b326 mm: filter based ... |
1769 |
classzone_idx = zone_idx(preferred_zone); |
9276b1bc9 [PATCH] memory pa... |
1770 |
zonelist_scan: |
7fb1d9fca [PATCH] mm: __all... |
1771 |
/* |
9276b1bc9 [PATCH] memory pa... |
1772 |
* Scan zonelist, looking for a zone with enough free. |
7fb1d9fca [PATCH] mm: __all... |
1773 1774 |
* See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ |
19770b326 mm: filter based ... |
1775 1776 |
for_each_zone_zonelist_nodemask(zone, z, zonelist, high_zoneidx, nodemask) { |
e5adfffc8 mm: use IS_ENABLE... |
1777 |
if (IS_ENABLED(CONFIG_NUMA) && zlc_active && |
9276b1bc9 [PATCH] memory pa... |
1778 1779 |
!zlc_zone_worth_trying(zonelist, z, allowednodes)) continue; |
7fb1d9fca [PATCH] mm: __all... |
1780 |
if ((alloc_flags & ALLOC_CPUSET) && |
02a0e53d8 [PATCH] cpuset: r... |
1781 |
!cpuset_zone_allowed_softwall(zone, gfp_mask)) |
cd38b115d mm: page allocato... |
1782 |
continue; |
a756cf590 mm: try to distri... |
1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 |
/* * When allocating a page cache page for writing, we * want to get it from a zone that is within its dirty * limit, such that no single zone holds more than its * proportional share of globally allowed dirty pages. * The dirty limits take into account the zone's * lowmem reserves and high watermark so that kswapd * should be able to balance it without having to * write pages from its LRU list. * * This may look like it could increase pressure on * lower zones by failing allocations in higher zones * before they are full. But the pages that do spill * over are limited as the lower zones are protected * by this very same mechanism. It should not become * a practical burden to them. * * XXX: For now, allow allocations to potentially * exceed the per-zone dirty limit in the slowpath * (ALLOC_WMARK_LOW unset) before going into reclaim, * which is important when on a NUMA setup the allowed * zones are together not big enough to reach the * global limit. The proper fix for these situations * will require awareness of zones in the * dirty-throttling and the flusher threads. */ if ((alloc_flags & ALLOC_WMARK_LOW) && (gfp_mask & __GFP_WRITE) && !zone_dirty_ok(zone)) goto this_zone_full; |
7fb1d9fca [PATCH] mm: __all... |
1812 |
|
418589663 page allocator: u... |
1813 |
BUILD_BUG_ON(ALLOC_NO_WATERMARKS < NR_WMARK); |
7fb1d9fca [PATCH] mm: __all... |
1814 |
if (!(alloc_flags & ALLOC_NO_WATERMARKS)) { |
3148890bf [PATCH] mm: __all... |
1815 |
unsigned long mark; |
fa5e084e4 vmscan: do not un... |
1816 |
int ret; |
418589663 page allocator: u... |
1817 |
mark = zone->watermark[alloc_flags & ALLOC_WMARK_MASK]; |
fa5e084e4 vmscan: do not un... |
1818 1819 1820 |
if (zone_watermark_ok(zone, order, mark, classzone_idx, alloc_flags)) goto try_this_zone; |
e5adfffc8 mm: use IS_ENABLE... |
1821 1822 |
if (IS_ENABLED(CONFIG_NUMA) && !did_zlc_setup && nr_online_nodes > 1) { |
cd38b115d mm: page allocato... |
1823 1824 1825 1826 1827 1828 1829 1830 1831 |
/* * we do zlc_setup if there are multiple nodes * and before considering the first zone allowed * by the cpuset. */ allowednodes = zlc_setup(zonelist, alloc_flags); zlc_active = 1; did_zlc_setup = 1; } |
957f822a0 mm, numa: reclaim... |
1832 1833 |
if (zone_reclaim_mode == 0 || !zone_allows_reclaim(preferred_zone, zone)) |
fa5e084e4 vmscan: do not un... |
1834 |
goto this_zone_full; |
cd38b115d mm: page allocato... |
1835 1836 1837 1838 |
/* * As we may have just activated ZLC, check if the first * eligible zone has failed zone_reclaim recently. */ |
e5adfffc8 mm: use IS_ENABLE... |
1839 |
if (IS_ENABLED(CONFIG_NUMA) && zlc_active && |
cd38b115d mm: page allocato... |
1840 1841 |
!zlc_zone_worth_trying(zonelist, z, allowednodes)) continue; |
fa5e084e4 vmscan: do not un... |
1842 1843 1844 1845 |
ret = zone_reclaim(zone, gfp_mask, order); switch (ret) { case ZONE_RECLAIM_NOSCAN: /* did not scan */ |
cd38b115d mm: page allocato... |
1846 |
continue; |
fa5e084e4 vmscan: do not un... |
1847 1848 |
case ZONE_RECLAIM_FULL: /* scanned but unreclaimable */ |
cd38b115d mm: page allocato... |
1849 |
continue; |
fa5e084e4 vmscan: do not un... |
1850 1851 |
default: /* did we reclaim enough */ |
fed2719e7 mm: page_alloc: a... |
1852 |
if (zone_watermark_ok(zone, order, mark, |
fa5e084e4 vmscan: do not un... |
1853 |
classzone_idx, alloc_flags)) |
fed2719e7 mm: page_alloc: a... |
1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 |
goto try_this_zone; /* * Failed to reclaim enough to meet watermark. * Only mark the zone full if checking the min * watermark or if we failed to reclaim just * 1<<order pages or else the page allocator * fastpath will prematurely mark zones full * when the watermark is between the low and * min watermarks. */ if (((alloc_flags & ALLOC_WMARK_MASK) == ALLOC_WMARK_MIN) || ret == ZONE_RECLAIM_SOME) |
9276b1bc9 [PATCH] memory pa... |
1867 |
goto this_zone_full; |
fed2719e7 mm: page_alloc: a... |
1868 1869 |
continue; |
0798e5193 [PATCH] memory pa... |
1870 |
} |
7fb1d9fca [PATCH] mm: __all... |
1871 |
} |
fa5e084e4 vmscan: do not un... |
1872 |
try_this_zone: |
3dd282669 page allocator: c... |
1873 1874 |
page = buffered_rmqueue(preferred_zone, zone, order, gfp_mask, migratetype); |
0798e5193 [PATCH] memory pa... |
1875 |
if (page) |
7fb1d9fca [PATCH] mm: __all... |
1876 |
break; |
9276b1bc9 [PATCH] memory pa... |
1877 |
this_zone_full: |
e5adfffc8 mm: use IS_ENABLE... |
1878 |
if (IS_ENABLED(CONFIG_NUMA)) |
9276b1bc9 [PATCH] memory pa... |
1879 |
zlc_mark_zone_full(zonelist, z); |
54a6eb5c4 mm: use two zonel... |
1880 |
} |
9276b1bc9 [PATCH] memory pa... |
1881 |
|
e5adfffc8 mm: use IS_ENABLE... |
1882 |
if (unlikely(IS_ENABLED(CONFIG_NUMA) && page == NULL && zlc_active)) { |
9276b1bc9 [PATCH] memory pa... |
1883 1884 1885 1886 |
/* Disable zlc cache for second zonelist scan */ zlc_active = 0; goto zonelist_scan; } |
b121186ab mm: correct page-... |
1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 |
if (page) /* * page->pfmemalloc is set when ALLOC_NO_WATERMARKS was * necessary to allocate the page. The expectation is * that the caller is taking steps that will free more * memory. The caller should avoid the page being used * for !PFMEMALLOC purposes. */ page->pfmemalloc = !!(alloc_flags & ALLOC_NO_WATERMARKS); |
7fb1d9fca [PATCH] mm: __all... |
1897 |
return page; |
753ee7289 [PATCH] VM: early... |
1898 |
} |
29423e77c oom: suppress sho... |
1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 |
/* * Large machines with many possible nodes should not always dump per-node * meminfo in irq context. */ static inline bool should_suppress_show_mem(void) { bool ret = false; #if NODES_SHIFT > 8 ret = in_interrupt(); #endif return ret; } |
a238ab5b0 mm: break out pag... |
1912 1913 1914 1915 1916 1917 |
static DEFINE_RATELIMIT_STATE(nopage_rs, DEFAULT_RATELIMIT_INTERVAL, DEFAULT_RATELIMIT_BURST); void warn_alloc_failed(gfp_t gfp_mask, int order, const char *fmt, ...) { |
a238ab5b0 mm: break out pag... |
1918 |
unsigned int filter = SHOW_MEM_FILTER_NODES; |
c0a32fc5a mm: more intensiv... |
1919 1920 |
if ((gfp_mask & __GFP_NOWARN) || !__ratelimit(&nopage_rs) || debug_guardpage_minorder() > 0) |
a238ab5b0 mm: break out pag... |
1921 1922 1923 |
return; /* |
4b59e6c47 mm, show_mem: sup... |
1924 1925 1926 1927 1928 1929 1930 |
* Walking all memory to count page types is very expensive and should * be inhibited in non-blockable contexts. */ if (!(gfp_mask & __GFP_WAIT)) filter |= SHOW_MEM_FILTER_PAGE_COUNT; /* |
a238ab5b0 mm: break out pag... |
1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 |
* This documents exceptions given to allocations in certain * contexts that are allowed to allocate outside current's set * of allowed nodes. */ if (!(gfp_mask & __GFP_NOMEMALLOC)) if (test_thread_flag(TIF_MEMDIE) || (current->flags & (PF_MEMALLOC | PF_EXITING))) filter &= ~SHOW_MEM_FILTER_NODES; if (in_interrupt() || !(gfp_mask & __GFP_WAIT)) filter &= ~SHOW_MEM_FILTER_NODES; if (fmt) { |
3ee9a4f08 mm: neaten warn_a... |
1943 1944 |
struct va_format vaf; va_list args; |
a238ab5b0 mm: break out pag... |
1945 |
va_start(args, fmt); |
3ee9a4f08 mm: neaten warn_a... |
1946 1947 1948 1949 1950 |
vaf.fmt = fmt; vaf.va = &args; pr_warn("%pV", &vaf); |
a238ab5b0 mm: break out pag... |
1951 1952 |
va_end(args); } |
3ee9a4f08 mm: neaten warn_a... |
1953 1954 1955 |
pr_warn("%s: page allocation failure: order:%d, mode:0x%x ", current->comm, order, gfp_mask); |
a238ab5b0 mm: break out pag... |
1956 1957 1958 1959 1960 |
dump_stack(); if (!should_suppress_show_mem()) show_mem(filter); } |
11e33f6a5 page allocator: b... |
1961 1962 |
static inline int should_alloc_retry(gfp_t gfp_mask, unsigned int order, |
f90ac3982 mm: avoid liveloc... |
1963 |
unsigned long did_some_progress, |
11e33f6a5 page allocator: b... |
1964 |
unsigned long pages_reclaimed) |
1da177e4c Linux-2.6.12-rc2 |
1965 |
{ |
11e33f6a5 page allocator: b... |
1966 1967 1968 |
/* Do not loop if specifically requested */ if (gfp_mask & __GFP_NORETRY) return 0; |
1da177e4c Linux-2.6.12-rc2 |
1969 |
|
f90ac3982 mm: avoid liveloc... |
1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 |
/* Always retry if specifically requested */ if (gfp_mask & __GFP_NOFAIL) return 1; /* * Suspend converts GFP_KERNEL to __GFP_WAIT which can prevent reclaim * making forward progress without invoking OOM. Suspend also disables * storage devices so kswapd will not help. Bail if we are suspending. */ if (!did_some_progress && pm_suspended_storage()) return 0; |
11e33f6a5 page allocator: b... |
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 |
/* * In this implementation, order <= PAGE_ALLOC_COSTLY_ORDER * means __GFP_NOFAIL, but that may not be true in other * implementations. */ if (order <= PAGE_ALLOC_COSTLY_ORDER) return 1; /* * For order > PAGE_ALLOC_COSTLY_ORDER, if __GFP_REPEAT is * specified, then we retry until we no longer reclaim any pages * (above), or we've reclaimed an order of pages at least as * large as the allocation's order. In both cases, if the * allocation still fails, we stop retrying. */ if (gfp_mask & __GFP_REPEAT && pages_reclaimed < (1 << order)) return 1; |
cf40bd16f lockdep: annotate... |
1998 |
|
11e33f6a5 page allocator: b... |
1999 2000 |
return 0; } |
933e312e7 [PATCH] fault-inj... |
2001 |
|
11e33f6a5 page allocator: b... |
2002 2003 2004 |
static inline struct page * __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, |
3dd282669 page allocator: c... |
2005 2006 |
nodemask_t *nodemask, struct zone *preferred_zone, int migratetype) |
11e33f6a5 page allocator: b... |
2007 2008 2009 2010 |
{ struct page *page; /* Acquire the OOM killer lock for the zones in zonelist */ |
ff321feac mm: rename try_se... |
2011 |
if (!try_set_zonelist_oom(zonelist, gfp_mask)) { |
11e33f6a5 page allocator: b... |
2012 |
schedule_timeout_uninterruptible(1); |
1da177e4c Linux-2.6.12-rc2 |
2013 2014 |
return NULL; } |
6b1de9161 [PATCH] VM: fix z... |
2015 |
|
11e33f6a5 page allocator: b... |
2016 2017 2018 2019 2020 2021 2022 |
/* * Go through the zonelist yet one more time, keep very high watermark * here, this is only to catch a parallel oom killing, we must fail if * we're still under heavy pressure. */ page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, nodemask, order, zonelist, high_zoneidx, |
5117f45d1 page allocator: c... |
2023 |
ALLOC_WMARK_HIGH|ALLOC_CPUSET, |
3dd282669 page allocator: c... |
2024 |
preferred_zone, migratetype); |
7fb1d9fca [PATCH] mm: __all... |
2025 |
if (page) |
11e33f6a5 page allocator: b... |
2026 |
goto out; |
4365a5676 oom-kill: fix NUM... |
2027 2028 2029 2030 |
if (!(gfp_mask & __GFP_NOFAIL)) { /* The OOM killer will not help higher order allocs */ if (order > PAGE_ALLOC_COSTLY_ORDER) goto out; |
03668b3ce oom: avoid oom ki... |
2031 2032 2033 |
/* The OOM killer does not needlessly kill tasks for lowmem */ if (high_zoneidx < ZONE_NORMAL) goto out; |
4365a5676 oom-kill: fix NUM... |
2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 |
/* * GFP_THISNODE contains __GFP_NORETRY and we never hit this. * Sanity check for bare calls of __GFP_THISNODE, not real OOM. * The caller should handle page allocation failure by itself if * it specifies __GFP_THISNODE. * Note: Hugepage uses it but will hit PAGE_ALLOC_COSTLY_ORDER. */ if (gfp_mask & __GFP_THISNODE) goto out; } |
11e33f6a5 page allocator: b... |
2044 |
/* Exhausted what can be done so it's blamo time */ |
08ab9b10d mm, oom: force oo... |
2045 |
out_of_memory(zonelist, gfp_mask, order, nodemask, false); |
11e33f6a5 page allocator: b... |
2046 2047 2048 2049 2050 |
out: clear_zonelist_oom(zonelist, gfp_mask); return page; } |
56de7263f mm: compaction: d... |
2051 2052 2053 2054 2055 2056 |
#ifdef CONFIG_COMPACTION /* Try memory compaction for high-order allocations before reclaim */ static struct page * __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone, |
66199712e mm: page allocato... |
2057 |
int migratetype, bool sync_migration, |
c67fe3752 mm: compaction: A... |
2058 |
bool *contended_compaction, bool *deferred_compaction, |
66199712e mm: page allocato... |
2059 |
unsigned long *did_some_progress) |
56de7263f mm: compaction: d... |
2060 |
{ |
66199712e mm: page allocato... |
2061 |
if (!order) |
56de7263f mm: compaction: d... |
2062 |
return NULL; |
aff622495 vmscan: only defe... |
2063 |
if (compaction_deferred(preferred_zone, order)) { |
66199712e mm: page allocato... |
2064 2065 2066 |
*deferred_compaction = true; return NULL; } |
c06b1fca1 mm/page_alloc.c: ... |
2067 |
current->flags |= PF_MEMALLOC; |
56de7263f mm: compaction: d... |
2068 |
*did_some_progress = try_to_compact_pages(zonelist, order, gfp_mask, |
c67fe3752 mm: compaction: A... |
2069 |
nodemask, sync_migration, |
8fb74b9fb mm: compaction: p... |
2070 |
contended_compaction); |
c06b1fca1 mm/page_alloc.c: ... |
2071 |
current->flags &= ~PF_MEMALLOC; |
56de7263f mm: compaction: d... |
2072 |
|
1fb3f8ca0 mm: compaction: c... |
2073 |
if (*did_some_progress != COMPACT_SKIPPED) { |
8fb74b9fb mm: compaction: p... |
2074 |
struct page *page; |
56de7263f mm: compaction: d... |
2075 2076 2077 2078 2079 2080 |
/* Page migration frees to the PCP lists but we want merging */ drain_pages(get_cpu()); put_cpu(); page = get_page_from_freelist(gfp_mask, nodemask, order, zonelist, high_zoneidx, |
cfd19c5a9 mm: only set page... |
2081 2082 |
alloc_flags & ~ALLOC_NO_WATERMARKS, preferred_zone, migratetype); |
56de7263f mm: compaction: d... |
2083 |
if (page) { |
62997027c mm: compaction: c... |
2084 |
preferred_zone->compact_blockskip_flush = false; |
4f92e2586 mm: compaction: d... |
2085 2086 |
preferred_zone->compact_considered = 0; preferred_zone->compact_defer_shift = 0; |
aff622495 vmscan: only defe... |
2087 2088 |
if (order >= preferred_zone->compact_order_failed) preferred_zone->compact_order_failed = order + 1; |
56de7263f mm: compaction: d... |
2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 |
count_vm_event(COMPACTSUCCESS); return page; } /* * It's bad if compaction run occurs and fails. * The most likely reason is that pages exist, * but not enough to satisfy watermarks. */ count_vm_event(COMPACTFAIL); |
66199712e mm: page allocato... |
2099 2100 2101 2102 2103 2104 |
/* * As async compaction considers a subset of pageblocks, only * defer if the failure was a sync compaction failure. */ if (sync_migration) |
aff622495 vmscan: only defe... |
2105 |
defer_compaction(preferred_zone, order); |
56de7263f mm: compaction: d... |
2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 |
cond_resched(); } return NULL; } #else static inline struct page * __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone, |
66199712e mm: page allocato... |
2117 |
int migratetype, bool sync_migration, |
c67fe3752 mm: compaction: A... |
2118 |
bool *contended_compaction, bool *deferred_compaction, |
66199712e mm: page allocato... |
2119 |
unsigned long *did_some_progress) |
56de7263f mm: compaction: d... |
2120 2121 2122 2123 |
{ return NULL; } #endif /* CONFIG_COMPACTION */ |
bba907108 mm: extract recla... |
2124 2125 2126 2127 |
/* Perform direct synchronous page reclaim */ static int __perform_reclaim(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, nodemask_t *nodemask) |
11e33f6a5 page allocator: b... |
2128 |
{ |
11e33f6a5 page allocator: b... |
2129 |
struct reclaim_state reclaim_state; |
bba907108 mm: extract recla... |
2130 |
int progress; |
11e33f6a5 page allocator: b... |
2131 2132 2133 2134 2135 |
cond_resched(); /* We now go into synchronous reclaim */ cpuset_memory_pressure_bump(); |
c06b1fca1 mm/page_alloc.c: ... |
2136 |
current->flags |= PF_MEMALLOC; |
11e33f6a5 page allocator: b... |
2137 2138 |
lockdep_set_current_reclaim_state(gfp_mask); reclaim_state.reclaimed_slab = 0; |
c06b1fca1 mm/page_alloc.c: ... |
2139 |
current->reclaim_state = &reclaim_state; |
11e33f6a5 page allocator: b... |
2140 |
|
bba907108 mm: extract recla... |
2141 |
progress = try_to_free_pages(zonelist, order, gfp_mask, nodemask); |
11e33f6a5 page allocator: b... |
2142 |
|
c06b1fca1 mm/page_alloc.c: ... |
2143 |
current->reclaim_state = NULL; |
11e33f6a5 page allocator: b... |
2144 |
lockdep_clear_current_reclaim_state(); |
c06b1fca1 mm/page_alloc.c: ... |
2145 |
current->flags &= ~PF_MEMALLOC; |
11e33f6a5 page allocator: b... |
2146 2147 |
cond_resched(); |
bba907108 mm: extract recla... |
2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 |
return progress; } /* The really slow allocator path where we enter direct reclaim */ static inline struct page * __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone, int migratetype, unsigned long *did_some_progress) { struct page *page = NULL; bool drained = false; *did_some_progress = __perform_reclaim(gfp_mask, order, zonelist, nodemask); |
9ee493ce0 mm: page allocato... |
2163 2164 |
if (unlikely(!(*did_some_progress))) return NULL; |
11e33f6a5 page allocator: b... |
2165 |
|
76d3fbf8f mm: page allocato... |
2166 |
/* After successful reclaim, reconsider all zones for allocation */ |
e5adfffc8 mm: use IS_ENABLE... |
2167 |
if (IS_ENABLED(CONFIG_NUMA)) |
76d3fbf8f mm: page allocato... |
2168 |
zlc_clear_zones_full(zonelist); |
9ee493ce0 mm: page allocato... |
2169 2170 |
retry: page = get_page_from_freelist(gfp_mask, nodemask, order, |
5117f45d1 page allocator: c... |
2171 |
zonelist, high_zoneidx, |
cfd19c5a9 mm: only set page... |
2172 2173 |
alloc_flags & ~ALLOC_NO_WATERMARKS, preferred_zone, migratetype); |
9ee493ce0 mm: page allocato... |
2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 |
/* * If an allocation failed after direct reclaim, it could be because * pages are pinned on the per-cpu lists. Drain them and try again */ if (!page && !drained) { drain_all_pages(); drained = true; goto retry; } |
11e33f6a5 page allocator: b... |
2184 2185 |
return page; } |
1da177e4c Linux-2.6.12-rc2 |
2186 |
/* |
11e33f6a5 page allocator: b... |
2187 2188 |
* This is called in the allocator slow-path if the allocation request is of * sufficient urgency to ignore watermarks and take other desperate measures |
1da177e4c Linux-2.6.12-rc2 |
2189 |
*/ |
11e33f6a5 page allocator: b... |
2190 2191 2192 |
static inline struct page * __alloc_pages_high_priority(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, |
3dd282669 page allocator: c... |
2193 2194 |
nodemask_t *nodemask, struct zone *preferred_zone, int migratetype) |
11e33f6a5 page allocator: b... |
2195 2196 2197 2198 2199 |
{ struct page *page; do { page = get_page_from_freelist(gfp_mask, nodemask, order, |
5117f45d1 page allocator: c... |
2200 |
zonelist, high_zoneidx, ALLOC_NO_WATERMARKS, |
3dd282669 page allocator: c... |
2201 |
preferred_zone, migratetype); |
11e33f6a5 page allocator: b... |
2202 2203 |
if (!page && gfp_mask & __GFP_NOFAIL) |
0e093d997 writeback: do not... |
2204 |
wait_iff_congested(preferred_zone, BLK_RW_ASYNC, HZ/50); |
11e33f6a5 page allocator: b... |
2205 2206 2207 2208 2209 2210 2211 |
} while (!page && (gfp_mask & __GFP_NOFAIL)); return page; } static inline void wake_all_kswapd(unsigned int order, struct zonelist *zonelist, |
995047488 mm: kswapd: stop ... |
2212 2213 |
enum zone_type high_zoneidx, enum zone_type classzone_idx) |
1da177e4c Linux-2.6.12-rc2 |
2214 |
{ |
dd1a239f6 mm: have zonelist... |
2215 2216 |
struct zoneref *z; struct zone *zone; |
1da177e4c Linux-2.6.12-rc2 |
2217 |
|
11e33f6a5 page allocator: b... |
2218 |
for_each_zone_zonelist(zone, z, zonelist, high_zoneidx) |
995047488 mm: kswapd: stop ... |
2219 |
wakeup_kswapd(zone, order, classzone_idx); |
11e33f6a5 page allocator: b... |
2220 |
} |
cf40bd16f lockdep: annotate... |
2221 |
|
341ce06f6 page allocator: c... |
2222 2223 2224 |
static inline int gfp_to_alloc_flags(gfp_t gfp_mask) { |
341ce06f6 page allocator: c... |
2225 2226 |
int alloc_flags = ALLOC_WMARK_MIN | ALLOC_CPUSET; const gfp_t wait = gfp_mask & __GFP_WAIT; |
1da177e4c Linux-2.6.12-rc2 |
2227 |
|
a56f57ff9 page allocator: r... |
2228 |
/* __GFP_HIGH is assumed to be the same as ALLOC_HIGH to save a branch. */ |
e6223a3b1 mm: add casts to/... |
2229 |
BUILD_BUG_ON(__GFP_HIGH != (__force gfp_t) ALLOC_HIGH); |
933e312e7 [PATCH] fault-inj... |
2230 |
|
341ce06f6 page allocator: c... |
2231 2232 2233 2234 2235 2236 |
/* * The caller may dip into page reserves a bit more if the caller * cannot run direct reclaim, or if the caller has realtime scheduling * policy or is asking for __GFP_HIGH memory. GFP_ATOMIC requests will * set both ALLOC_HARDER (!wait) and ALLOC_HIGH (__GFP_HIGH). */ |
e6223a3b1 mm: add casts to/... |
2237 |
alloc_flags |= (__force int) (gfp_mask & __GFP_HIGH); |
1da177e4c Linux-2.6.12-rc2 |
2238 |
|
341ce06f6 page allocator: c... |
2239 |
if (!wait) { |
5c3240d92 thp: don't alloc ... |
2240 2241 2242 2243 2244 2245 |
/* * Not worth trying to allocate harder for * __GFP_NOMEMALLOC even if it can't schedule. */ if (!(gfp_mask & __GFP_NOMEMALLOC)) alloc_flags |= ALLOC_HARDER; |
523b94585 Memoryless nodes:... |
2246 |
/* |
341ce06f6 page allocator: c... |
2247 2248 |
* Ignore cpuset if GFP_ATOMIC (!wait) rather than fail alloc. * See also cpuset_zone_allowed() comment in kernel/cpuset.c. |
523b94585 Memoryless nodes:... |
2249 |
*/ |
341ce06f6 page allocator: c... |
2250 |
alloc_flags &= ~ALLOC_CPUSET; |
c06b1fca1 mm/page_alloc.c: ... |
2251 |
} else if (unlikely(rt_task(current)) && !in_interrupt()) |
341ce06f6 page allocator: c... |
2252 |
alloc_flags |= ALLOC_HARDER; |
b37f1dd0f mm: introduce __G... |
2253 2254 2255 |
if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) { if (gfp_mask & __GFP_MEMALLOC) alloc_flags |= ALLOC_NO_WATERMARKS; |
907aed48f mm: allow PF_MEMA... |
2256 2257 2258 2259 2260 |
else if (in_serving_softirq() && (current->flags & PF_MEMALLOC)) alloc_flags |= ALLOC_NO_WATERMARKS; else if (!in_interrupt() && ((current->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE)))) |
341ce06f6 page allocator: c... |
2261 |
alloc_flags |= ALLOC_NO_WATERMARKS; |
1da177e4c Linux-2.6.12-rc2 |
2262 |
} |
d95ea5d18 cma: fix watermar... |
2263 2264 2265 2266 |
#ifdef CONFIG_CMA if (allocflags_to_migratetype(gfp_mask) == MIGRATE_MOVABLE) alloc_flags |= ALLOC_CMA; #endif |
341ce06f6 page allocator: c... |
2267 2268 |
return alloc_flags; } |
072bb0aa5 mm: sl[au]b: add ... |
2269 2270 |
bool gfp_pfmemalloc_allowed(gfp_t gfp_mask) { |
b37f1dd0f mm: introduce __G... |
2271 |
return !!(gfp_to_alloc_flags(gfp_mask) & ALLOC_NO_WATERMARKS); |
072bb0aa5 mm: sl[au]b: add ... |
2272 |
} |
11e33f6a5 page allocator: b... |
2273 2274 2275 |
static inline struct page * __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, |
3dd282669 page allocator: c... |
2276 2277 |
nodemask_t *nodemask, struct zone *preferred_zone, int migratetype) |
11e33f6a5 page allocator: b... |
2278 2279 2280 2281 2282 2283 |
{ const gfp_t wait = gfp_mask & __GFP_WAIT; struct page *page = NULL; int alloc_flags; unsigned long pages_reclaimed = 0; unsigned long did_some_progress; |
77f1fe6b0 mm: migration: al... |
2284 |
bool sync_migration = false; |
66199712e mm: page allocato... |
2285 |
bool deferred_compaction = false; |
c67fe3752 mm: compaction: A... |
2286 |
bool contended_compaction = false; |
1da177e4c Linux-2.6.12-rc2 |
2287 |
|
952f3b51b [PATCH] GFP_THISN... |
2288 |
/* |
72807a74c page allocator: s... |
2289 2290 2291 2292 2293 |
* In the slowpath, we sanity check order to avoid ever trying to * reclaim >= MAX_ORDER areas which will never succeed. Callers may * be using allocators in order of preference for an area that is * too large. */ |
1fc28b70f page-allocator: a... |
2294 2295 |
if (order >= MAX_ORDER) { WARN_ON_ONCE(!(gfp_mask & __GFP_NOWARN)); |
72807a74c page allocator: s... |
2296 |
return NULL; |
1fc28b70f page-allocator: a... |
2297 |
} |
1da177e4c Linux-2.6.12-rc2 |
2298 |
|
952f3b51b [PATCH] GFP_THISN... |
2299 2300 2301 2302 2303 2304 2305 2306 |
/* * GFP_THISNODE (meaning __GFP_THISNODE, __GFP_NORETRY and * __GFP_NOWARN set) should not cause reclaim since the subsystem * (f.e. slab) using GFP_THISNODE may choose to trigger reclaim * using a larger set of nodes after it has established that the * allowed per node queues are empty and that nodes are * over allocated. */ |
e5adfffc8 mm: use IS_ENABLE... |
2307 2308 |
if (IS_ENABLED(CONFIG_NUMA) && (gfp_mask & GFP_THISNODE) == GFP_THISNODE) |
952f3b51b [PATCH] GFP_THISN... |
2309 |
goto nopage; |
cc4a68514 page allocator: a... |
2310 |
restart: |
caf491916 Revert "revert "R... |
2311 2312 2313 |
if (!(gfp_mask & __GFP_NO_KSWAPD)) wake_all_kswapd(order, zonelist, high_zoneidx, zone_idx(preferred_zone)); |
1da177e4c Linux-2.6.12-rc2 |
2314 |
|
9bf2229f8 [PATCH] cpusets: ... |
2315 |
/* |
7fb1d9fca [PATCH] mm: __all... |
2316 2317 2318 |
* OK, we're below the kswapd watermark and have kicked background * reclaim. Now things get more complex, so set up alloc_flags according * to how we want to proceed. |
9bf2229f8 [PATCH] cpusets: ... |
2319 |
*/ |
341ce06f6 page allocator: c... |
2320 |
alloc_flags = gfp_to_alloc_flags(gfp_mask); |
1da177e4c Linux-2.6.12-rc2 |
2321 |
|
f33261d75 mm: fix deferred ... |
2322 2323 2324 2325 2326 2327 2328 |
/* * Find the true preferred zone if the allocation is unconstrained by * cpusets. */ if (!(alloc_flags & ALLOC_CPUSET) && !nodemask) first_zones_zonelist(zonelist, high_zoneidx, NULL, &preferred_zone); |
cfa54a0fc mm/page_alloc.c: ... |
2329 |
rebalance: |
341ce06f6 page allocator: c... |
2330 |
/* This is the last chance, in general, before the goto nopage. */ |
19770b326 mm: filter based ... |
2331 |
page = get_page_from_freelist(gfp_mask, nodemask, order, zonelist, |
341ce06f6 page allocator: c... |
2332 2333 |
high_zoneidx, alloc_flags & ~ALLOC_NO_WATERMARKS, preferred_zone, migratetype); |
7fb1d9fca [PATCH] mm: __all... |
2334 2335 |
if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
2336 |
|
11e33f6a5 page allocator: b... |
2337 |
/* Allocate without watermarks if the context allows */ |
341ce06f6 page allocator: c... |
2338 |
if (alloc_flags & ALLOC_NO_WATERMARKS) { |
183f6371a mm: ignore mempol... |
2339 2340 2341 2342 2343 2344 |
/* * Ignore mempolicies if ALLOC_NO_WATERMARKS on the grounds * the allocation is high priority and these type of * allocations are system rather than user orientated */ zonelist = node_zonelist(numa_node_id(), gfp_mask); |
341ce06f6 page allocator: c... |
2345 2346 2347 |
page = __alloc_pages_high_priority(gfp_mask, order, zonelist, high_zoneidx, nodemask, preferred_zone, migratetype); |
cfd19c5a9 mm: only set page... |
2348 |
if (page) { |
341ce06f6 page allocator: c... |
2349 |
goto got_pg; |
cfd19c5a9 mm: only set page... |
2350 |
} |
1da177e4c Linux-2.6.12-rc2 |
2351 2352 2353 2354 2355 |
} /* Atomic allocations - we can't balance anything */ if (!wait) goto nopage; |
341ce06f6 page allocator: c... |
2356 |
/* Avoid recursion of direct reclaim */ |
c06b1fca1 mm/page_alloc.c: ... |
2357 |
if (current->flags & PF_MEMALLOC) |
341ce06f6 page allocator: c... |
2358 |
goto nopage; |
6583bb64f mm: avoid endless... |
2359 2360 2361 |
/* Avoid allocations with no watermarks from looping endlessly */ if (test_thread_flag(TIF_MEMDIE) && !(gfp_mask & __GFP_NOFAIL)) goto nopage; |
77f1fe6b0 mm: migration: al... |
2362 2363 2364 2365 |
/* * Try direct compaction. The first pass is asynchronous. Subsequent * attempts after direct reclaim are synchronous */ |
56de7263f mm: compaction: d... |
2366 2367 2368 2369 |
page = __alloc_pages_direct_compact(gfp_mask, order, zonelist, high_zoneidx, nodemask, alloc_flags, preferred_zone, |
66199712e mm: page allocato... |
2370 |
migratetype, sync_migration, |
c67fe3752 mm: compaction: A... |
2371 |
&contended_compaction, |
66199712e mm: page allocato... |
2372 2373 |
&deferred_compaction, &did_some_progress); |
56de7263f mm: compaction: d... |
2374 2375 |
if (page) goto got_pg; |
c6a140bf1 mm/compaction: re... |
2376 |
sync_migration = true; |
56de7263f mm: compaction: d... |
2377 |
|
31f8d42d4 Revert "mm: avoid... |
2378 2379 2380 2381 2382 2383 2384 |
/* * If compaction is deferred for high-order allocations, it is because * sync compaction recently failed. In this is the case and the caller * requested a movable allocation that does not heavily disrupt the * system then fail the allocation instead of entering direct reclaim. */ if ((deferred_compaction || contended_compaction) && |
caf491916 Revert "revert "R... |
2385 |
(gfp_mask & __GFP_NO_KSWAPD)) |
31f8d42d4 Revert "mm: avoid... |
2386 |
goto nopage; |
66199712e mm: page allocato... |
2387 |
|
11e33f6a5 page allocator: b... |
2388 2389 2390 2391 |
/* Try direct reclaim and then allocating */ page = __alloc_pages_direct_reclaim(gfp_mask, order, zonelist, high_zoneidx, nodemask, |
5117f45d1 page allocator: c... |
2392 |
alloc_flags, preferred_zone, |
3dd282669 page allocator: c... |
2393 |
migratetype, &did_some_progress); |
11e33f6a5 page allocator: b... |
2394 2395 |
if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
2396 |
|
e33c3b5e1 cpusets: update m... |
2397 |
/* |
11e33f6a5 page allocator: b... |
2398 2399 |
* If we failed to make any progress reclaiming, then we are * running out of options and have to consider going OOM |
e33c3b5e1 cpusets: update m... |
2400 |
*/ |
11e33f6a5 page allocator: b... |
2401 2402 |
if (!did_some_progress) { if ((gfp_mask & __GFP_FS) && !(gfp_mask & __GFP_NORETRY)) { |
7f33d49a2 mm, PM/Freezer: D... |
2403 2404 |
if (oom_killer_disabled) goto nopage; |
29fd66d28 mm, coredump: fai... |
2405 2406 2407 2408 |
/* Coredumps can quickly deplete all memory reserves */ if ((current->flags & PF_DUMPCORE) && !(gfp_mask & __GFP_NOFAIL)) goto nopage; |
11e33f6a5 page allocator: b... |
2409 2410 |
page = __alloc_pages_may_oom(gfp_mask, order, zonelist, high_zoneidx, |
3dd282669 page allocator: c... |
2411 2412 |
nodemask, preferred_zone, migratetype); |
11e33f6a5 page allocator: b... |
2413 2414 |
if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
2415 |
|
03668b3ce oom: avoid oom ki... |
2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 |
if (!(gfp_mask & __GFP_NOFAIL)) { /* * The oom killer is not called for high-order * allocations that may fail, so if no progress * is being made, there are no other options and * retrying is unlikely to help. */ if (order > PAGE_ALLOC_COSTLY_ORDER) goto nopage; /* * The oom killer is not called for lowmem * allocations to prevent needlessly killing * innocent tasks. */ if (high_zoneidx < ZONE_NORMAL) goto nopage; } |
e2c55dc87 Drain per-cpu lis... |
2433 |
|
ff0ceb9de oom: serialize ou... |
2434 2435 |
goto restart; } |
1da177e4c Linux-2.6.12-rc2 |
2436 |
} |
11e33f6a5 page allocator: b... |
2437 |
/* Check if we should retry the allocation */ |
a41f24ea9 page allocator: s... |
2438 |
pages_reclaimed += did_some_progress; |
f90ac3982 mm: avoid liveloc... |
2439 2440 |
if (should_alloc_retry(gfp_mask, order, did_some_progress, pages_reclaimed)) { |
11e33f6a5 page allocator: b... |
2441 |
/* Wait for some write requests to complete then retry */ |
0e093d997 writeback: do not... |
2442 |
wait_iff_congested(preferred_zone, BLK_RW_ASYNC, HZ/50); |
1da177e4c Linux-2.6.12-rc2 |
2443 |
goto rebalance; |
3e7d34497 mm: vmscan: recla... |
2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 |
} else { /* * High-order allocations do not necessarily loop after * direct reclaim and reclaim/compaction depends on compaction * being called after reclaim so call directly if necessary */ page = __alloc_pages_direct_compact(gfp_mask, order, zonelist, high_zoneidx, nodemask, alloc_flags, preferred_zone, |
66199712e mm: page allocato... |
2454 |
migratetype, sync_migration, |
c67fe3752 mm: compaction: A... |
2455 |
&contended_compaction, |
66199712e mm: page allocato... |
2456 2457 |
&deferred_compaction, &did_some_progress); |
3e7d34497 mm: vmscan: recla... |
2458 2459 |
if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
2460 2461 2462 |
} nopage: |
a238ab5b0 mm: break out pag... |
2463 |
warn_alloc_failed(gfp_mask, order, NULL); |
b1eeab676 kmemcheck: add ho... |
2464 |
return page; |
1da177e4c Linux-2.6.12-rc2 |
2465 |
got_pg: |
b1eeab676 kmemcheck: add ho... |
2466 2467 |
if (kmemcheck_enabled) kmemcheck_pagealloc_alloc(page, order, gfp_mask); |
11e33f6a5 page allocator: b... |
2468 |
|
072bb0aa5 mm: sl[au]b: add ... |
2469 |
return page; |
1da177e4c Linux-2.6.12-rc2 |
2470 |
} |
11e33f6a5 page allocator: b... |
2471 2472 2473 2474 2475 2476 2477 2478 2479 |
/* * This is the 'heart' of the zoned buddy allocator. */ struct page * __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, nodemask_t *nodemask) { enum zone_type high_zoneidx = gfp_zone(gfp_mask); |
5117f45d1 page allocator: c... |
2480 |
struct zone *preferred_zone; |
cc9a6c877 cpuset: mm: reduc... |
2481 |
struct page *page = NULL; |
3dd282669 page allocator: c... |
2482 |
int migratetype = allocflags_to_migratetype(gfp_mask); |
cc9a6c877 cpuset: mm: reduc... |
2483 |
unsigned int cpuset_mems_cookie; |
d95ea5d18 cma: fix watermar... |
2484 |
int alloc_flags = ALLOC_WMARK_LOW|ALLOC_CPUSET; |
6a1a0d3b6 mm: allocate kern... |
2485 |
struct mem_cgroup *memcg = NULL; |
11e33f6a5 page allocator: b... |
2486 |
|
dcce284a2 mm: Extend gfp ma... |
2487 |
gfp_mask &= gfp_allowed_mask; |
11e33f6a5 page allocator: b... |
2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 |
lockdep_trace_alloc(gfp_mask); might_sleep_if(gfp_mask & __GFP_WAIT); if (should_fail_alloc_page(gfp_mask, order)) return NULL; /* * Check the zones suitable for the gfp_mask contain at least one * valid zone. It's possible to have an empty zonelist as a result * of GFP_THISNODE and a memoryless node */ if (unlikely(!zonelist->_zonerefs->zone)) return NULL; |
6a1a0d3b6 mm: allocate kern... |
2502 2503 2504 2505 2506 2507 |
/* * Will only have any effect when __GFP_KMEMCG is set. This is * verified in the (always inline) callee */ if (!memcg_kmem_newpage_charge(gfp_mask, &memcg, order)) return NULL; |
cc9a6c877 cpuset: mm: reduc... |
2508 2509 |
retry_cpuset: cpuset_mems_cookie = get_mems_allowed(); |
5117f45d1 page allocator: c... |
2510 |
/* The preferred zone is used for statistics later */ |
f33261d75 mm: fix deferred ... |
2511 2512 2513 |
first_zones_zonelist(zonelist, high_zoneidx, nodemask ? : &cpuset_current_mems_allowed, &preferred_zone); |
cc9a6c877 cpuset: mm: reduc... |
2514 2515 |
if (!preferred_zone) goto out; |
5117f45d1 page allocator: c... |
2516 |
|
d95ea5d18 cma: fix watermar... |
2517 2518 2519 2520 |
#ifdef CONFIG_CMA if (allocflags_to_migratetype(gfp_mask) == MIGRATE_MOVABLE) alloc_flags |= ALLOC_CMA; #endif |
5117f45d1 page allocator: c... |
2521 |
/* First allocation attempt */ |
11e33f6a5 page allocator: b... |
2522 |
page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, nodemask, order, |
d95ea5d18 cma: fix watermar... |
2523 |
zonelist, high_zoneidx, alloc_flags, |
3dd282669 page allocator: c... |
2524 |
preferred_zone, migratetype); |
21caf2fc1 mm: teach mm by c... |
2525 2526 2527 2528 2529 2530 2531 |
if (unlikely(!page)) { /* * Runtime PM, block IO and its error handling path * can deadlock because I/O on the device might not * complete. */ gfp_mask = memalloc_noio_flags(gfp_mask); |
11e33f6a5 page allocator: b... |
2532 |
page = __alloc_pages_slowpath(gfp_mask, order, |
5117f45d1 page allocator: c... |
2533 |
zonelist, high_zoneidx, nodemask, |
3dd282669 page allocator: c... |
2534 |
preferred_zone, migratetype); |
21caf2fc1 mm: teach mm by c... |
2535 |
} |
11e33f6a5 page allocator: b... |
2536 |
|
4b4f278c0 tracing, page-all... |
2537 |
trace_mm_page_alloc(page, order, gfp_mask, migratetype); |
cc9a6c877 cpuset: mm: reduc... |
2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 |
out: /* * When updating a task's mems_allowed, it is possible to race with * parallel threads in such a way that an allocation can fail while * the mask is being updated. If a page allocation is about to fail, * check if the cpuset changed during allocation and if so, retry. */ if (unlikely(!put_mems_allowed(cpuset_mems_cookie) && !page)) goto retry_cpuset; |
6a1a0d3b6 mm: allocate kern... |
2548 |
memcg_kmem_commit_charge(page, memcg, order); |
11e33f6a5 page allocator: b... |
2549 |
return page; |
1da177e4c Linux-2.6.12-rc2 |
2550 |
} |
d239171e4 page allocator: r... |
2551 |
EXPORT_SYMBOL(__alloc_pages_nodemask); |
1da177e4c Linux-2.6.12-rc2 |
2552 2553 2554 2555 |
/* * Common helper functions. */ |
920c7a5d0 mm: remove fastca... |
2556 |
unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order) |
1da177e4c Linux-2.6.12-rc2 |
2557 |
{ |
945a11136 mm: add gfp mask ... |
2558 2559 2560 2561 2562 2563 2564 |
struct page *page; /* * __get_free_pages() returns a 32-bit address, which cannot represent * a highmem page */ VM_BUG_ON((gfp_mask & __GFP_HIGHMEM) != 0); |
1da177e4c Linux-2.6.12-rc2 |
2565 2566 2567 2568 2569 |
page = alloc_pages(gfp_mask, order); if (!page) return 0; return (unsigned long) page_address(page); } |
1da177e4c Linux-2.6.12-rc2 |
2570 |
EXPORT_SYMBOL(__get_free_pages); |
920c7a5d0 mm: remove fastca... |
2571 |
unsigned long get_zeroed_page(gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
2572 |
{ |
945a11136 mm: add gfp mask ... |
2573 |
return __get_free_pages(gfp_mask | __GFP_ZERO, 0); |
1da177e4c Linux-2.6.12-rc2 |
2574 |
} |
1da177e4c Linux-2.6.12-rc2 |
2575 |
EXPORT_SYMBOL(get_zeroed_page); |
920c7a5d0 mm: remove fastca... |
2576 |
void __free_pages(struct page *page, unsigned int order) |
1da177e4c Linux-2.6.12-rc2 |
2577 |
{ |
b5810039a [PATCH] core remo... |
2578 |
if (put_page_testzero(page)) { |
1da177e4c Linux-2.6.12-rc2 |
2579 |
if (order == 0) |
fc91668ea mm: remove free_h... |
2580 |
free_hot_cold_page(page, 0); |
1da177e4c Linux-2.6.12-rc2 |
2581 2582 2583 2584 2585 2586 |
else __free_pages_ok(page, order); } } EXPORT_SYMBOL(__free_pages); |
920c7a5d0 mm: remove fastca... |
2587 |
void free_pages(unsigned long addr, unsigned int order) |
1da177e4c Linux-2.6.12-rc2 |
2588 2589 |
{ if (addr != 0) { |
725d704ec [PATCH] mm: VM_BU... |
2590 |
VM_BUG_ON(!virt_addr_valid((void *)addr)); |
1da177e4c Linux-2.6.12-rc2 |
2591 2592 2593 2594 2595 |
__free_pages(virt_to_page((void *)addr), order); } } EXPORT_SYMBOL(free_pages); |
6a1a0d3b6 mm: allocate kern... |
2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 |
/* * __free_memcg_kmem_pages and free_memcg_kmem_pages will free * pages allocated with __GFP_KMEMCG. * * Those pages are accounted to a particular memcg, embedded in the * corresponding page_cgroup. To avoid adding a hit in the allocator to search * for that information only to find out that it is NULL for users who have no * interest in that whatsoever, we provide these functions. * * The caller knows better which flags it relies on. */ void __free_memcg_kmem_pages(struct page *page, unsigned int order) { memcg_kmem_uncharge_pages(page, order); __free_pages(page, order); } void free_memcg_kmem_pages(unsigned long addr, unsigned int order) { if (addr != 0) { VM_BUG_ON(!virt_addr_valid((void *)addr)); __free_memcg_kmem_pages(virt_to_page((void *)addr), order); } } |
ee85c2e14 mm: add alloc_pag... |
2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 |
static void *make_alloc_exact(unsigned long addr, unsigned order, size_t size) { if (addr) { unsigned long alloc_end = addr + (PAGE_SIZE << order); unsigned long used = addr + PAGE_ALIGN(size); split_page(virt_to_page((void *)addr), order); while (used < alloc_end) { free_page(used); used += PAGE_SIZE; } } return (void *)addr; } |
2be0ffe2b mm: add alloc_pag... |
2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 |
/** * alloc_pages_exact - allocate an exact number physically-contiguous pages. * @size: the number of bytes to allocate * @gfp_mask: GFP flags for the allocation * * This function is similar to alloc_pages(), except that it allocates the * minimum number of pages to satisfy the request. alloc_pages() can only * allocate memory in power-of-two pages. * * This function is also limited by MAX_ORDER. * * Memory allocated by this function must be released by free_pages_exact(). */ void *alloc_pages_exact(size_t size, gfp_t gfp_mask) { unsigned int order = get_order(size); unsigned long addr; addr = __get_free_pages(gfp_mask, order); |
ee85c2e14 mm: add alloc_pag... |
2653 |
return make_alloc_exact(addr, order, size); |
2be0ffe2b mm: add alloc_pag... |
2654 2655 2656 2657 |
} EXPORT_SYMBOL(alloc_pages_exact); /** |
ee85c2e14 mm: add alloc_pag... |
2658 2659 |
* alloc_pages_exact_nid - allocate an exact number of physically-contiguous * pages on a node. |
b5e6ab589 mm: fix kernel-do... |
2660 |
* @nid: the preferred node ID where memory should be allocated |
ee85c2e14 mm: add alloc_pag... |
2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 |
* @size: the number of bytes to allocate * @gfp_mask: GFP flags for the allocation * * Like alloc_pages_exact(), but try to allocate on node nid first before falling * back. * Note this is not alloc_pages_exact_node() which allocates on a specific node, * but is not exact. */ void *alloc_pages_exact_nid(int nid, size_t size, gfp_t gfp_mask) { unsigned order = get_order(size); struct page *p = alloc_pages_node(nid, gfp_mask, order); if (!p) return NULL; return make_alloc_exact((unsigned long)page_address(p), order, size); } EXPORT_SYMBOL(alloc_pages_exact_nid); /** |
2be0ffe2b mm: add alloc_pag... |
2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 |
* free_pages_exact - release memory allocated via alloc_pages_exact() * @virt: the value returned by alloc_pages_exact. * @size: size of allocation, same value as passed to alloc_pages_exact(). * * Release the memory allocated by a previous call to alloc_pages_exact. */ void free_pages_exact(void *virt, size_t size) { unsigned long addr = (unsigned long)virt; unsigned long end = addr + PAGE_ALIGN(size); while (addr < end) { free_page(addr); addr += PAGE_SIZE; } } EXPORT_SYMBOL(free_pages_exact); |
e0fb58152 mm: accurately do... |
2697 2698 2699 2700 2701 2702 2703 2704 2705 |
/** * nr_free_zone_pages - count number of pages beyond high watermark * @offset: The zone index of the highest zone * * nr_free_zone_pages() counts the number of counts pages which are beyond the * high watermark within all zones at or below a given zone index. For each * zone, the number of pages is calculated as: * present_pages - high_pages */ |
ebec3862f mm: fix return ty... |
2706 |
static unsigned long nr_free_zone_pages(int offset) |
1da177e4c Linux-2.6.12-rc2 |
2707 |
{ |
dd1a239f6 mm: have zonelist... |
2708 |
struct zoneref *z; |
54a6eb5c4 mm: use two zonel... |
2709 |
struct zone *zone; |
e310fd432 [PATCH] Fix NUMA ... |
2710 |
/* Just pick one node, since fallback list is circular */ |
ebec3862f mm: fix return ty... |
2711 |
unsigned long sum = 0; |
1da177e4c Linux-2.6.12-rc2 |
2712 |
|
0e88460da mm: introduce nod... |
2713 |
struct zonelist *zonelist = node_zonelist(numa_node_id(), GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
2714 |
|
54a6eb5c4 mm: use two zonel... |
2715 |
for_each_zone_zonelist(zone, z, zonelist, offset) { |
b40da0494 mm: use zone->pre... |
2716 |
unsigned long size = zone->managed_pages; |
418589663 page allocator: u... |
2717 |
unsigned long high = high_wmark_pages(zone); |
e310fd432 [PATCH] Fix NUMA ... |
2718 2719 |
if (size > high) sum += size - high; |
1da177e4c Linux-2.6.12-rc2 |
2720 2721 2722 2723 |
} return sum; } |
e0fb58152 mm: accurately do... |
2724 2725 2726 2727 2728 |
/** * nr_free_buffer_pages - count number of pages beyond high watermark * * nr_free_buffer_pages() counts the number of pages which are beyond the high * watermark within ZONE_DMA and ZONE_NORMAL. |
1da177e4c Linux-2.6.12-rc2 |
2729 |
*/ |
ebec3862f mm: fix return ty... |
2730 |
unsigned long nr_free_buffer_pages(void) |
1da177e4c Linux-2.6.12-rc2 |
2731 |
{ |
af4ca457e [PATCH] gfp_t: in... |
2732 |
return nr_free_zone_pages(gfp_zone(GFP_USER)); |
1da177e4c Linux-2.6.12-rc2 |
2733 |
} |
c2f1a551d knfsd: nfsd4: var... |
2734 |
EXPORT_SYMBOL_GPL(nr_free_buffer_pages); |
1da177e4c Linux-2.6.12-rc2 |
2735 |
|
e0fb58152 mm: accurately do... |
2736 2737 2738 2739 2740 |
/** * nr_free_pagecache_pages - count number of pages beyond high watermark * * nr_free_pagecache_pages() counts the number of pages which are beyond the * high watermark within all zones. |
1da177e4c Linux-2.6.12-rc2 |
2741 |
*/ |
ebec3862f mm: fix return ty... |
2742 |
unsigned long nr_free_pagecache_pages(void) |
1da177e4c Linux-2.6.12-rc2 |
2743 |
{ |
2a1e274ac Create the ZONE_M... |
2744 |
return nr_free_zone_pages(gfp_zone(GFP_HIGHUSER_MOVABLE)); |
1da177e4c Linux-2.6.12-rc2 |
2745 |
} |
08e0f6a97 [PATCH] Add NUMA_... |
2746 2747 |
static inline void show_node(struct zone *zone) |
1da177e4c Linux-2.6.12-rc2 |
2748 |
{ |
e5adfffc8 mm: use IS_ENABLE... |
2749 |
if (IS_ENABLED(CONFIG_NUMA)) |
25ba77c14 [PATCH] numa node... |
2750 |
printk("Node %d ", zone_to_nid(zone)); |
1da177e4c Linux-2.6.12-rc2 |
2751 |
} |
1da177e4c Linux-2.6.12-rc2 |
2752 |
|
1da177e4c Linux-2.6.12-rc2 |
2753 2754 2755 2756 |
void si_meminfo(struct sysinfo *val) { val->totalram = totalram_pages; val->sharedram = 0; |
d23ad4232 [PATCH] Use ZVC f... |
2757 |
val->freeram = global_page_state(NR_FREE_PAGES); |
1da177e4c Linux-2.6.12-rc2 |
2758 |
val->bufferram = nr_blockdev_pages(); |
1da177e4c Linux-2.6.12-rc2 |
2759 2760 |
val->totalhigh = totalhigh_pages; val->freehigh = nr_free_highpages(); |
1da177e4c Linux-2.6.12-rc2 |
2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 |
val->mem_unit = PAGE_SIZE; } EXPORT_SYMBOL(si_meminfo); #ifdef CONFIG_NUMA void si_meminfo_node(struct sysinfo *val, int nid) { pg_data_t *pgdat = NODE_DATA(nid); val->totalram = pgdat->node_present_pages; |
d23ad4232 [PATCH] Use ZVC f... |
2772 |
val->freeram = node_page_state(nid, NR_FREE_PAGES); |
98d2b0ebd [PATCH] reduce MA... |
2773 |
#ifdef CONFIG_HIGHMEM |
b40da0494 mm: use zone->pre... |
2774 |
val->totalhigh = pgdat->node_zones[ZONE_HIGHMEM].managed_pages; |
d23ad4232 [PATCH] Use ZVC f... |
2775 2776 |
val->freehigh = zone_page_state(&pgdat->node_zones[ZONE_HIGHMEM], NR_FREE_PAGES); |
98d2b0ebd [PATCH] reduce MA... |
2777 2778 2779 2780 |
#else val->totalhigh = 0; val->freehigh = 0; #endif |
1da177e4c Linux-2.6.12-rc2 |
2781 2782 2783 |
val->mem_unit = PAGE_SIZE; } #endif |
ddd588b5d oom: suppress nod... |
2784 |
/* |
7bf02ea22 arch, mm: filter ... |
2785 2786 |
* Determine whether the node should be displayed or not, depending on whether * SHOW_MEM_FILTER_NODES was passed to show_free_areas(). |
ddd588b5d oom: suppress nod... |
2787 |
*/ |
7bf02ea22 arch, mm: filter ... |
2788 |
bool skip_free_areas_node(unsigned int flags, int nid) |
ddd588b5d oom: suppress nod... |
2789 2790 |
{ bool ret = false; |
cc9a6c877 cpuset: mm: reduc... |
2791 |
unsigned int cpuset_mems_cookie; |
ddd588b5d oom: suppress nod... |
2792 2793 2794 |
if (!(flags & SHOW_MEM_FILTER_NODES)) goto out; |
cc9a6c877 cpuset: mm: reduc... |
2795 2796 2797 2798 |
do { cpuset_mems_cookie = get_mems_allowed(); ret = !node_isset(nid, cpuset_current_mems_allowed); } while (!put_mems_allowed(cpuset_mems_cookie)); |
ddd588b5d oom: suppress nod... |
2799 2800 2801 |
out: return ret; } |
1da177e4c Linux-2.6.12-rc2 |
2802 |
#define K(x) ((x) << (PAGE_SHIFT-10)) |
377e4f167 mm: show migratio... |
2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 |
static void show_migration_types(unsigned char type) { static const char types[MIGRATE_TYPES] = { [MIGRATE_UNMOVABLE] = 'U', [MIGRATE_RECLAIMABLE] = 'E', [MIGRATE_MOVABLE] = 'M', [MIGRATE_RESERVE] = 'R', #ifdef CONFIG_CMA [MIGRATE_CMA] = 'C', #endif |
194159fbc mm: remove MIGRAT... |
2813 |
#ifdef CONFIG_MEMORY_ISOLATION |
377e4f167 mm: show migratio... |
2814 |
[MIGRATE_ISOLATE] = 'I', |
194159fbc mm: remove MIGRAT... |
2815 |
#endif |
377e4f167 mm: show migratio... |
2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 |
}; char tmp[MIGRATE_TYPES + 1]; char *p = tmp; int i; for (i = 0; i < MIGRATE_TYPES; i++) { if (type & (1 << i)) *p++ = types[i]; } *p = '\0'; printk("(%s) ", tmp); } |
1da177e4c Linux-2.6.12-rc2 |
2829 2830 2831 2832 |
/* * Show free area list (used inside shift_scroll-lock stuff) * We also calculate the percentage fragmentation. We do this by counting the * memory on each free list with the exception of the first item on the list. |
ddd588b5d oom: suppress nod... |
2833 2834 |
* Suppresses nodes that are not allowed by current's cpuset if * SHOW_MEM_FILTER_NODES is passed. |
1da177e4c Linux-2.6.12-rc2 |
2835 |
*/ |
7bf02ea22 arch, mm: filter ... |
2836 |
void show_free_areas(unsigned int filter) |
1da177e4c Linux-2.6.12-rc2 |
2837 |
{ |
c72419138 [PATCH] Condense ... |
2838 |
int cpu; |
1da177e4c Linux-2.6.12-rc2 |
2839 |
struct zone *zone; |
ee99c71c5 mm: introduce for... |
2840 |
for_each_populated_zone(zone) { |
7bf02ea22 arch, mm: filter ... |
2841 |
if (skip_free_areas_node(filter, zone_to_nid(zone))) |
ddd588b5d oom: suppress nod... |
2842 |
continue; |
c72419138 [PATCH] Condense ... |
2843 2844 2845 |
show_node(zone); printk("%s per-cpu: ", zone->name); |
1da177e4c Linux-2.6.12-rc2 |
2846 |
|
6b482c677 [PATCH] Don't pri... |
2847 |
for_each_online_cpu(cpu) { |
1da177e4c Linux-2.6.12-rc2 |
2848 |
struct per_cpu_pageset *pageset; |
99dcc3e5a this_cpu: Page al... |
2849 |
pageset = per_cpu_ptr(zone->pageset, cpu); |
1da177e4c Linux-2.6.12-rc2 |
2850 |
|
3dfa5721f Page allocator: g... |
2851 2852 2853 2854 |
printk("CPU %4d: hi:%5d, btch:%4d usd:%4d ", cpu, pageset->pcp.high, pageset->pcp.batch, pageset->pcp.count); |
1da177e4c Linux-2.6.12-rc2 |
2855 2856 |
} } |
a731286de mm: vmstat: add i... |
2857 2858 2859 2860 |
printk("active_anon:%lu inactive_anon:%lu isolated_anon:%lu " " active_file:%lu inactive_file:%lu isolated_file:%lu " |
7b854121e Unevictable LRU P... |
2861 |
" unevictable:%lu" |
b76146ed1 revert "mm: oom a... |
2862 2863 |
" dirty:%lu writeback:%lu unstable:%lu " |
3701b0332 mm: show_free_are... |
2864 2865 |
" free:%lu slab_reclaimable:%lu slab_unreclaimable:%lu " |
d1ce749a0 cma: count free C... |
2866 2867 2868 2869 |
" mapped:%lu shmem:%lu pagetables:%lu bounce:%lu " " free_cma:%lu ", |
4f98a2fee vmscan: split LRU... |
2870 |
global_page_state(NR_ACTIVE_ANON), |
4f98a2fee vmscan: split LRU... |
2871 |
global_page_state(NR_INACTIVE_ANON), |
a731286de mm: vmstat: add i... |
2872 2873 |
global_page_state(NR_ISOLATED_ANON), global_page_state(NR_ACTIVE_FILE), |
4f98a2fee vmscan: split LRU... |
2874 |
global_page_state(NR_INACTIVE_FILE), |
a731286de mm: vmstat: add i... |
2875 |
global_page_state(NR_ISOLATED_FILE), |
7b854121e Unevictable LRU P... |
2876 |
global_page_state(NR_UNEVICTABLE), |
b1e7a8fd8 [PATCH] zoned vm ... |
2877 |
global_page_state(NR_FILE_DIRTY), |
ce866b34a [PATCH] zoned vm ... |
2878 |
global_page_state(NR_WRITEBACK), |
fd39fc856 [PATCH] zoned vm ... |
2879 |
global_page_state(NR_UNSTABLE_NFS), |
d23ad4232 [PATCH] Use ZVC f... |
2880 |
global_page_state(NR_FREE_PAGES), |
3701b0332 mm: show_free_are... |
2881 2882 |
global_page_state(NR_SLAB_RECLAIMABLE), global_page_state(NR_SLAB_UNRECLAIMABLE), |
65ba55f50 [PATCH] zoned vm ... |
2883 |
global_page_state(NR_FILE_MAPPED), |
4b02108ac mm: oom analysis:... |
2884 |
global_page_state(NR_SHMEM), |
a25700a53 [PATCH] mm: show ... |
2885 |
global_page_state(NR_PAGETABLE), |
d1ce749a0 cma: count free C... |
2886 2887 |
global_page_state(NR_BOUNCE), global_page_state(NR_FREE_CMA_PAGES)); |
1da177e4c Linux-2.6.12-rc2 |
2888 |
|
ee99c71c5 mm: introduce for... |
2889 |
for_each_populated_zone(zone) { |
1da177e4c Linux-2.6.12-rc2 |
2890 |
int i; |
7bf02ea22 arch, mm: filter ... |
2891 |
if (skip_free_areas_node(filter, zone_to_nid(zone))) |
ddd588b5d oom: suppress nod... |
2892 |
continue; |
1da177e4c Linux-2.6.12-rc2 |
2893 2894 2895 2896 2897 2898 |
show_node(zone); printk("%s" " free:%lukB" " min:%lukB" " low:%lukB" " high:%lukB" |
4f98a2fee vmscan: split LRU... |
2899 2900 2901 2902 |
" active_anon:%lukB" " inactive_anon:%lukB" " active_file:%lukB" " inactive_file:%lukB" |
7b854121e Unevictable LRU P... |
2903 |
" unevictable:%lukB" |
a731286de mm: vmstat: add i... |
2904 2905 |
" isolated(anon):%lukB" " isolated(file):%lukB" |
1da177e4c Linux-2.6.12-rc2 |
2906 |
" present:%lukB" |
9feedc9d8 mm: introduce new... |
2907 |
" managed:%lukB" |
4a0aa73f1 mm: oom analysis:... |
2908 2909 2910 2911 |
" mlocked:%lukB" " dirty:%lukB" " writeback:%lukB" " mapped:%lukB" |
4b02108ac mm: oom analysis:... |
2912 |
" shmem:%lukB" |
4a0aa73f1 mm: oom analysis:... |
2913 2914 |
" slab_reclaimable:%lukB" " slab_unreclaimable:%lukB" |
c6a7f5728 mm: oom analysis:... |
2915 |
" kernel_stack:%lukB" |
4a0aa73f1 mm: oom analysis:... |
2916 2917 2918 |
" pagetables:%lukB" " unstable:%lukB" " bounce:%lukB" |
d1ce749a0 cma: count free C... |
2919 |
" free_cma:%lukB" |
4a0aa73f1 mm: oom analysis:... |
2920 |
" writeback_tmp:%lukB" |
1da177e4c Linux-2.6.12-rc2 |
2921 2922 2923 2924 2925 |
" pages_scanned:%lu" " all_unreclaimable? %s" " ", zone->name, |
88f5acf88 mm: page allocato... |
2926 |
K(zone_page_state(zone, NR_FREE_PAGES)), |
418589663 page allocator: u... |
2927 2928 2929 |
K(min_wmark_pages(zone)), K(low_wmark_pages(zone)), K(high_wmark_pages(zone)), |
4f98a2fee vmscan: split LRU... |
2930 2931 2932 2933 |
K(zone_page_state(zone, NR_ACTIVE_ANON)), K(zone_page_state(zone, NR_INACTIVE_ANON)), K(zone_page_state(zone, NR_ACTIVE_FILE)), K(zone_page_state(zone, NR_INACTIVE_FILE)), |
7b854121e Unevictable LRU P... |
2934 |
K(zone_page_state(zone, NR_UNEVICTABLE)), |
a731286de mm: vmstat: add i... |
2935 2936 |
K(zone_page_state(zone, NR_ISOLATED_ANON)), K(zone_page_state(zone, NR_ISOLATED_FILE)), |
1da177e4c Linux-2.6.12-rc2 |
2937 |
K(zone->present_pages), |
9feedc9d8 mm: introduce new... |
2938 |
K(zone->managed_pages), |
4a0aa73f1 mm: oom analysis:... |
2939 2940 2941 2942 |
K(zone_page_state(zone, NR_MLOCK)), K(zone_page_state(zone, NR_FILE_DIRTY)), K(zone_page_state(zone, NR_WRITEBACK)), K(zone_page_state(zone, NR_FILE_MAPPED)), |
4b02108ac mm: oom analysis:... |
2943 |
K(zone_page_state(zone, NR_SHMEM)), |
4a0aa73f1 mm: oom analysis:... |
2944 2945 |
K(zone_page_state(zone, NR_SLAB_RECLAIMABLE)), K(zone_page_state(zone, NR_SLAB_UNRECLAIMABLE)), |
c6a7f5728 mm: oom analysis:... |
2946 2947 |
zone_page_state(zone, NR_KERNEL_STACK) * THREAD_SIZE / 1024, |
4a0aa73f1 mm: oom analysis:... |
2948 2949 2950 |
K(zone_page_state(zone, NR_PAGETABLE)), K(zone_page_state(zone, NR_UNSTABLE_NFS)), K(zone_page_state(zone, NR_BOUNCE)), |
d1ce749a0 cma: count free C... |
2951 |
K(zone_page_state(zone, NR_FREE_CMA_PAGES)), |
4a0aa73f1 mm: oom analysis:... |
2952 |
K(zone_page_state(zone, NR_WRITEBACK_TEMP)), |
1da177e4c Linux-2.6.12-rc2 |
2953 |
zone->pages_scanned, |
93e4a89a8 mm: restore zone-... |
2954 |
(zone->all_unreclaimable ? "yes" : "no") |
1da177e4c Linux-2.6.12-rc2 |
2955 2956 2957 2958 2959 2960 2961 |
); printk("lowmem_reserve[]:"); for (i = 0; i < MAX_NR_ZONES; i++) printk(" %lu", zone->lowmem_reserve[i]); printk(" "); } |
ee99c71c5 mm: introduce for... |
2962 |
for_each_populated_zone(zone) { |
8f9de51a4 [PATCH] printk() ... |
2963 |
unsigned long nr[MAX_ORDER], flags, order, total = 0; |
377e4f167 mm: show migratio... |
2964 |
unsigned char types[MAX_ORDER]; |
1da177e4c Linux-2.6.12-rc2 |
2965 |
|
7bf02ea22 arch, mm: filter ... |
2966 |
if (skip_free_areas_node(filter, zone_to_nid(zone))) |
ddd588b5d oom: suppress nod... |
2967 |
continue; |
1da177e4c Linux-2.6.12-rc2 |
2968 2969 |
show_node(zone); printk("%s: ", zone->name); |
1da177e4c Linux-2.6.12-rc2 |
2970 2971 2972 |
spin_lock_irqsave(&zone->lock, flags); for (order = 0; order < MAX_ORDER; order++) { |
377e4f167 mm: show migratio... |
2973 2974 2975 2976 |
struct free_area *area = &zone->free_area[order]; int type; nr[order] = area->nr_free; |
8f9de51a4 [PATCH] printk() ... |
2977 |
total += nr[order] << order; |
377e4f167 mm: show migratio... |
2978 2979 2980 2981 2982 2983 |
types[order] = 0; for (type = 0; type < MIGRATE_TYPES; type++) { if (!list_empty(&area->free_list[type])) types[order] |= 1 << type; } |
1da177e4c Linux-2.6.12-rc2 |
2984 2985 |
} spin_unlock_irqrestore(&zone->lock, flags); |
377e4f167 mm: show migratio... |
2986 |
for (order = 0; order < MAX_ORDER; order++) { |
8f9de51a4 [PATCH] printk() ... |
2987 |
printk("%lu*%lukB ", nr[order], K(1UL) << order); |
377e4f167 mm: show migratio... |
2988 2989 2990 |
if (nr[order]) show_migration_types(types[order]); } |
1da177e4c Linux-2.6.12-rc2 |
2991 2992 2993 |
printk("= %lukB ", K(total)); } |
949f7ec57 mm, hugetlb: incl... |
2994 |
hugetlb_show_meminfo(); |
e6f3602d2 Include count of ... |
2995 2996 |
printk("%ld total pagecache pages ", global_page_state(NR_FILE_PAGES)); |
1da177e4c Linux-2.6.12-rc2 |
2997 2998 |
show_swap_cache_info(); } |
19770b326 mm: filter based ... |
2999 3000 3001 3002 3003 |
static void zoneref_set_zone(struct zone *zone, struct zoneref *zoneref) { zoneref->zone = zone; zoneref->zone_idx = zone_idx(zone); } |
1da177e4c Linux-2.6.12-rc2 |
3004 3005 |
/* * Builds allocation fallback zone lists. |
1a93205bd [PATCH] mm: simpl... |
3006 3007 |
* * Add all populated zones of a node to the zonelist. |
1da177e4c Linux-2.6.12-rc2 |
3008 |
*/ |
f0c0b2b80 change zonelist o... |
3009 3010 |
static int build_zonelists_node(pg_data_t *pgdat, struct zonelist *zonelist, int nr_zones, enum zone_type zone_type) |
1da177e4c Linux-2.6.12-rc2 |
3011 |
{ |
1a93205bd [PATCH] mm: simpl... |
3012 |
struct zone *zone; |
98d2b0ebd [PATCH] reduce MA... |
3013 |
BUG_ON(zone_type >= MAX_NR_ZONES); |
2f6726e54 [PATCH] Apply typ... |
3014 |
zone_type++; |
02a68a5eb [PATCH] Fix zone ... |
3015 3016 |
do { |
2f6726e54 [PATCH] Apply typ... |
3017 |
zone_type--; |
070f80326 [PATCH] build_zon... |
3018 |
zone = pgdat->node_zones + zone_type; |
1a93205bd [PATCH] mm: simpl... |
3019 |
if (populated_zone(zone)) { |
dd1a239f6 mm: have zonelist... |
3020 3021 |
zoneref_set_zone(zone, &zonelist->_zonerefs[nr_zones++]); |
070f80326 [PATCH] build_zon... |
3022 |
check_highest_zone(zone_type); |
1da177e4c Linux-2.6.12-rc2 |
3023 |
} |
02a68a5eb [PATCH] Fix zone ... |
3024 |
|
2f6726e54 [PATCH] Apply typ... |
3025 |
} while (zone_type); |
070f80326 [PATCH] build_zon... |
3026 |
return nr_zones; |
1da177e4c Linux-2.6.12-rc2 |
3027 |
} |
f0c0b2b80 change zonelist o... |
3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 |
/* * zonelist_order: * 0 = automatic detection of better ordering. * 1 = order by ([node] distance, -zonetype) * 2 = order by (-zonetype, [node] distance) * * If not NUMA, ZONELIST_ORDER_ZONE and ZONELIST_ORDER_NODE will create * the same zonelist. So only NUMA can configure this param. */ #define ZONELIST_ORDER_DEFAULT 0 #define ZONELIST_ORDER_NODE 1 #define ZONELIST_ORDER_ZONE 2 /* zonelist order in the kernel. * set_zonelist_order() will set this to NODE or ZONE. */ static int current_zonelist_order = ZONELIST_ORDER_DEFAULT; static char zonelist_order_name[3][8] = {"Default", "Node", "Zone"}; |
1da177e4c Linux-2.6.12-rc2 |
3047 |
#ifdef CONFIG_NUMA |
f0c0b2b80 change zonelist o... |
3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 |
/* The value user specified ....changed by config */ static int user_zonelist_order = ZONELIST_ORDER_DEFAULT; /* string for sysctl */ #define NUMA_ZONELIST_ORDER_LEN 16 char numa_zonelist_order[16] = "default"; /* * interface for configure zonelist ordering. * command line option "numa_zonelist_order" * = "[dD]efault - default, automatic configuration. * = "[nN]ode - order by node locality, then by zone within node * = "[zZ]one - order by zone, then by locality within zone */ static int __parse_numa_zonelist_order(char *s) { if (*s == 'd' || *s == 'D') { user_zonelist_order = ZONELIST_ORDER_DEFAULT; } else if (*s == 'n' || *s == 'N') { user_zonelist_order = ZONELIST_ORDER_NODE; } else if (*s == 'z' || *s == 'Z') { user_zonelist_order = ZONELIST_ORDER_ZONE; } else { printk(KERN_WARNING "Ignoring invalid numa_zonelist_order value: " "%s ", s); return -EINVAL; } return 0; } static __init int setup_numa_zonelist_order(char *s) { |
ecb256f81 mm: set correct n... |
3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 |
int ret; if (!s) return 0; ret = __parse_numa_zonelist_order(s); if (ret == 0) strlcpy(numa_zonelist_order, s, NUMA_ZONELIST_ORDER_LEN); return ret; |
f0c0b2b80 change zonelist o... |
3092 3093 3094 3095 3096 3097 3098 |
} early_param("numa_zonelist_order", setup_numa_zonelist_order); /* * sysctl handler for numa_zonelist_order */ int numa_zonelist_order_handler(ctl_table *table, int write, |
8d65af789 sysctl: remove "s... |
3099 |
void __user *buffer, size_t *length, |
f0c0b2b80 change zonelist o... |
3100 3101 3102 3103 |
loff_t *ppos) { char saved_string[NUMA_ZONELIST_ORDER_LEN]; int ret; |
443c6f145 SYSCTL: Add a mut... |
3104 |
static DEFINE_MUTEX(zl_order_mutex); |
f0c0b2b80 change zonelist o... |
3105 |
|
443c6f145 SYSCTL: Add a mut... |
3106 |
mutex_lock(&zl_order_mutex); |
f0c0b2b80 change zonelist o... |
3107 |
if (write) |
443c6f145 SYSCTL: Add a mut... |
3108 |
strcpy(saved_string, (char*)table->data); |
8d65af789 sysctl: remove "s... |
3109 |
ret = proc_dostring(table, write, buffer, length, ppos); |
f0c0b2b80 change zonelist o... |
3110 |
if (ret) |
443c6f145 SYSCTL: Add a mut... |
3111 |
goto out; |
f0c0b2b80 change zonelist o... |
3112 3113 3114 3115 3116 3117 3118 3119 3120 |
if (write) { int oldval = user_zonelist_order; if (__parse_numa_zonelist_order((char*)table->data)) { /* * bogus value. restore saved string */ strncpy((char*)table->data, saved_string, NUMA_ZONELIST_ORDER_LEN); user_zonelist_order = oldval; |
4eaf3f643 mem-hotplug: fix ... |
3121 3122 |
} else if (oldval != user_zonelist_order) { mutex_lock(&zonelists_mutex); |
9adb62a5d mm/hotplug: corre... |
3123 |
build_all_zonelists(NULL, NULL); |
4eaf3f643 mem-hotplug: fix ... |
3124 3125 |
mutex_unlock(&zonelists_mutex); } |
f0c0b2b80 change zonelist o... |
3126 |
} |
443c6f145 SYSCTL: Add a mut... |
3127 3128 3129 |
out: mutex_unlock(&zl_order_mutex); return ret; |
f0c0b2b80 change zonelist o... |
3130 |
} |
62bc62a87 page allocator: u... |
3131 |
#define MAX_NODE_LOAD (nr_online_nodes) |
f0c0b2b80 change zonelist o... |
3132 |
static int node_load[MAX_NUMNODES]; |
1da177e4c Linux-2.6.12-rc2 |
3133 |
/** |
4dc3b16ba [PATCH] DocBook: ... |
3134 |
* find_next_best_node - find the next node that should appear in a given node's fallback list |
1da177e4c Linux-2.6.12-rc2 |
3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 |
* @node: node whose fallback list we're appending * @used_node_mask: nodemask_t of already used nodes * * We use a number of factors to determine which is the next node that should * appear on a given node's fallback list. The node should not have appeared * already in @node's fallback list, and it should be the next closest node * according to the distance array (which contains arbitrary distance values * from each node to each node in the system), and should also prefer nodes * with no CPUs, since presumably they'll have very little allocation pressure * on them otherwise. * It returns -1 if no node is found. */ |
f0c0b2b80 change zonelist o... |
3147 |
static int find_next_best_node(int node, nodemask_t *used_node_mask) |
1da177e4c Linux-2.6.12-rc2 |
3148 |
{ |
4cf808eb4 [PATCH] Handle ho... |
3149 |
int n, val; |
1da177e4c Linux-2.6.12-rc2 |
3150 |
int min_val = INT_MAX; |
00ef2d2f8 mm: use NUMA_NO_NODE |
3151 |
int best_node = NUMA_NO_NODE; |
a70f73028 cpumask: replace ... |
3152 |
const struct cpumask *tmp = cpumask_of_node(0); |
1da177e4c Linux-2.6.12-rc2 |
3153 |
|
4cf808eb4 [PATCH] Handle ho... |
3154 3155 3156 3157 3158 |
/* Use the local node if we haven't already */ if (!node_isset(node, *used_node_mask)) { node_set(node, *used_node_mask); return node; } |
1da177e4c Linux-2.6.12-rc2 |
3159 |
|
4b0ef1fe8 page_alloc: use N... |
3160 |
for_each_node_state(n, N_MEMORY) { |
1da177e4c Linux-2.6.12-rc2 |
3161 3162 3163 3164 |
/* Don't want a node to appear more than once */ if (node_isset(n, *used_node_mask)) continue; |
1da177e4c Linux-2.6.12-rc2 |
3165 3166 |
/* Use the distance array to find the distance */ val = node_distance(node, n); |
4cf808eb4 [PATCH] Handle ho... |
3167 3168 |
/* Penalize nodes under us ("prefer the next node") */ val += (n < node); |
1da177e4c Linux-2.6.12-rc2 |
3169 |
/* Give preference to headless and unused nodes */ |
a70f73028 cpumask: replace ... |
3170 3171 |
tmp = cpumask_of_node(n); if (!cpumask_empty(tmp)) |
1da177e4c Linux-2.6.12-rc2 |
3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 |
val += PENALTY_FOR_NODE_WITH_CPUS; /* Slight preference for less loaded node */ val *= (MAX_NODE_LOAD*MAX_NUMNODES); val += node_load[n]; if (val < min_val) { min_val = val; best_node = n; } } if (best_node >= 0) node_set(best_node, *used_node_mask); return best_node; } |
f0c0b2b80 change zonelist o... |
3189 3190 3191 3192 3193 3194 3195 |
/* * Build zonelists ordered by node and zones within node. * This results in maximum locality--normal zone overflows into local * DMA zone, if any--but risks exhausting DMA zone. */ static void build_zonelists_in_node_order(pg_data_t *pgdat, int node) |
1da177e4c Linux-2.6.12-rc2 |
3196 |
{ |
f0c0b2b80 change zonelist o... |
3197 |
int j; |
1da177e4c Linux-2.6.12-rc2 |
3198 |
struct zonelist *zonelist; |
f0c0b2b80 change zonelist o... |
3199 |
|
54a6eb5c4 mm: use two zonel... |
3200 |
zonelist = &pgdat->node_zonelists[0]; |
dd1a239f6 mm: have zonelist... |
3201 |
for (j = 0; zonelist->_zonerefs[j].zone != NULL; j++) |
54a6eb5c4 mm: use two zonel... |
3202 3203 3204 |
; j = build_zonelists_node(NODE_DATA(node), zonelist, j, MAX_NR_ZONES - 1); |
dd1a239f6 mm: have zonelist... |
3205 3206 |
zonelist->_zonerefs[j].zone = NULL; zonelist->_zonerefs[j].zone_idx = 0; |
f0c0b2b80 change zonelist o... |
3207 3208 3209 |
} /* |
523b94585 Memoryless nodes:... |
3210 3211 3212 3213 |
* Build gfp_thisnode zonelists */ static void build_thisnode_zonelists(pg_data_t *pgdat) { |
523b94585 Memoryless nodes:... |
3214 3215 |
int j; struct zonelist *zonelist; |
54a6eb5c4 mm: use two zonel... |
3216 3217 |
zonelist = &pgdat->node_zonelists[1]; j = build_zonelists_node(pgdat, zonelist, 0, MAX_NR_ZONES - 1); |
dd1a239f6 mm: have zonelist... |
3218 3219 |
zonelist->_zonerefs[j].zone = NULL; zonelist->_zonerefs[j].zone_idx = 0; |
523b94585 Memoryless nodes:... |
3220 3221 3222 |
} /* |
f0c0b2b80 change zonelist o... |
3223 3224 3225 3226 3227 3228 3229 3230 3231 |
* Build zonelists ordered by zone and nodes within zones. * This results in conserving DMA zone[s] until all Normal memory is * exhausted, but results in overflowing to remote node while memory * may still exist in local DMA zone. */ static int node_order[MAX_NUMNODES]; static void build_zonelists_in_zone_order(pg_data_t *pgdat, int nr_nodes) { |
f0c0b2b80 change zonelist o... |
3232 3233 3234 3235 |
int pos, j, node; int zone_type; /* needs to be signed */ struct zone *z; struct zonelist *zonelist; |
54a6eb5c4 mm: use two zonel... |
3236 3237 3238 3239 3240 3241 3242 |
zonelist = &pgdat->node_zonelists[0]; pos = 0; for (zone_type = MAX_NR_ZONES - 1; zone_type >= 0; zone_type--) { for (j = 0; j < nr_nodes; j++) { node = node_order[j]; z = &NODE_DATA(node)->node_zones[zone_type]; if (populated_zone(z)) { |
dd1a239f6 mm: have zonelist... |
3243 3244 |
zoneref_set_zone(z, &zonelist->_zonerefs[pos++]); |
54a6eb5c4 mm: use two zonel... |
3245 |
check_highest_zone(zone_type); |
f0c0b2b80 change zonelist o... |
3246 3247 |
} } |
f0c0b2b80 change zonelist o... |
3248 |
} |
dd1a239f6 mm: have zonelist... |
3249 3250 |
zonelist->_zonerefs[pos].zone = NULL; zonelist->_zonerefs[pos].zone_idx = 0; |
f0c0b2b80 change zonelist o... |
3251 3252 3253 3254 3255 3256 3257 3258 3259 |
} static int default_zonelist_order(void) { int nid, zone_type; unsigned long low_kmem_size,total_size; struct zone *z; int average_size; /* |
883931612 Fix typos in comm... |
3260 |
* ZONE_DMA and ZONE_DMA32 can be very small area in the system. |
f0c0b2b80 change zonelist o... |
3261 3262 |
* If they are really small and used heavily, the system can fall * into OOM very easily. |
e325c90ff mm: default to no... |
3263 |
* This function detect ZONE_DMA/DMA32 size and configures zone order. |
f0c0b2b80 change zonelist o... |
3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 |
*/ /* Is there ZONE_NORMAL ? (ex. ppc has only DMA zone..) */ low_kmem_size = 0; total_size = 0; for_each_online_node(nid) { for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++) { z = &NODE_DATA(nid)->node_zones[zone_type]; if (populated_zone(z)) { if (zone_type < ZONE_NORMAL) low_kmem_size += z->present_pages; total_size += z->present_pages; |
e325c90ff mm: default to no... |
3275 3276 3277 3278 3279 3280 3281 3282 3283 |
} else if (zone_type == ZONE_NORMAL) { /* * If any node has only lowmem, then node order * is preferred to allow kernel allocations * locally; otherwise, they can easily infringe * on other nodes when there is an abundance of * lowmem available to allocate from. */ return ZONELIST_ORDER_NODE; |
f0c0b2b80 change zonelist o... |
3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 |
} } } if (!low_kmem_size || /* there are no DMA area. */ low_kmem_size > total_size/2) /* DMA/DMA32 is big. */ return ZONELIST_ORDER_NODE; /* * look into each node's config. * If there is a node whose DMA/DMA32 memory is very big area on * local memory, NODE_ORDER may be suitable. */ |
37b07e416 memoryless nodes:... |
3295 |
average_size = total_size / |
4b0ef1fe8 page_alloc: use N... |
3296 |
(nodes_weight(node_states[N_MEMORY]) + 1); |
f0c0b2b80 change zonelist o... |
3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 |
for_each_online_node(nid) { low_kmem_size = 0; total_size = 0; for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++) { z = &NODE_DATA(nid)->node_zones[zone_type]; if (populated_zone(z)) { if (zone_type < ZONE_NORMAL) low_kmem_size += z->present_pages; total_size += z->present_pages; } } if (low_kmem_size && total_size > average_size && /* ignore small node */ low_kmem_size > total_size * 70/100) return ZONELIST_ORDER_NODE; } return ZONELIST_ORDER_ZONE; } static void set_zonelist_order(void) { if (user_zonelist_order == ZONELIST_ORDER_DEFAULT) current_zonelist_order = default_zonelist_order(); else current_zonelist_order = user_zonelist_order; } static void build_zonelists(pg_data_t *pgdat) { int j, node, load; enum zone_type i; |
1da177e4c Linux-2.6.12-rc2 |
3328 |
nodemask_t used_mask; |
f0c0b2b80 change zonelist o... |
3329 3330 3331 |
int local_node, prev_node; struct zonelist *zonelist; int order = current_zonelist_order; |
1da177e4c Linux-2.6.12-rc2 |
3332 3333 |
/* initialize zonelists */ |
523b94585 Memoryless nodes:... |
3334 |
for (i = 0; i < MAX_ZONELISTS; i++) { |
1da177e4c Linux-2.6.12-rc2 |
3335 |
zonelist = pgdat->node_zonelists + i; |
dd1a239f6 mm: have zonelist... |
3336 3337 |
zonelist->_zonerefs[0].zone = NULL; zonelist->_zonerefs[0].zone_idx = 0; |
1da177e4c Linux-2.6.12-rc2 |
3338 3339 3340 3341 |
} /* NUMA-aware ordering of nodes */ local_node = pgdat->node_id; |
62bc62a87 page allocator: u... |
3342 |
load = nr_online_nodes; |
1da177e4c Linux-2.6.12-rc2 |
3343 3344 |
prev_node = local_node; nodes_clear(used_mask); |
f0c0b2b80 change zonelist o... |
3345 |
|
f0c0b2b80 change zonelist o... |
3346 3347 |
memset(node_order, 0, sizeof(node_order)); j = 0; |
1da177e4c Linux-2.6.12-rc2 |
3348 3349 3350 3351 3352 3353 |
while ((node = find_next_best_node(local_node, &used_mask)) >= 0) { /* * We don't want to pressure a particular node. * So adding penalty to the first node in same * distance group to make it round-robin. */ |
957f822a0 mm, numa: reclaim... |
3354 3355 |
if (node_distance(local_node, node) != node_distance(local_node, prev_node)) |
f0c0b2b80 change zonelist o... |
3356 |
node_load[node] = load; |
1da177e4c Linux-2.6.12-rc2 |
3357 3358 |
prev_node = node; load--; |
f0c0b2b80 change zonelist o... |
3359 3360 3361 3362 3363 |
if (order == ZONELIST_ORDER_NODE) build_zonelists_in_node_order(pgdat, node); else node_order[j++] = node; /* remember order */ } |
1da177e4c Linux-2.6.12-rc2 |
3364 |
|
f0c0b2b80 change zonelist o... |
3365 3366 3367 |
if (order == ZONELIST_ORDER_ZONE) { /* calculate node order -- i.e., DMA last! */ build_zonelists_in_zone_order(pgdat, j); |
1da177e4c Linux-2.6.12-rc2 |
3368 |
} |
523b94585 Memoryless nodes:... |
3369 3370 |
build_thisnode_zonelists(pgdat); |
1da177e4c Linux-2.6.12-rc2 |
3371 |
} |
9276b1bc9 [PATCH] memory pa... |
3372 |
/* Construct the zonelist performance cache - see further mmzone.h */ |
f0c0b2b80 change zonelist o... |
3373 |
static void build_zonelist_cache(pg_data_t *pgdat) |
9276b1bc9 [PATCH] memory pa... |
3374 |
{ |
54a6eb5c4 mm: use two zonel... |
3375 3376 |
struct zonelist *zonelist; struct zonelist_cache *zlc; |
dd1a239f6 mm: have zonelist... |
3377 |
struct zoneref *z; |
9276b1bc9 [PATCH] memory pa... |
3378 |
|
54a6eb5c4 mm: use two zonel... |
3379 3380 3381 |
zonelist = &pgdat->node_zonelists[0]; zonelist->zlcache_ptr = zlc = &zonelist->zlcache; bitmap_zero(zlc->fullzones, MAX_ZONES_PER_ZONELIST); |
dd1a239f6 mm: have zonelist... |
3382 3383 |
for (z = zonelist->_zonerefs; z->zone; z++) zlc->z_to_n[z - zonelist->_zonerefs] = zonelist_node_idx(z); |
9276b1bc9 [PATCH] memory pa... |
3384 |
} |
7aac78988 numa: introduce n... |
3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 |
#ifdef CONFIG_HAVE_MEMORYLESS_NODES /* * Return node id of node used for "local" allocations. * I.e., first node id of first zone in arg node's generic zonelist. * Used for initializing percpu 'numa_mem', which is used primarily * for kernel allocations, so use GFP_KERNEL flags to locate zonelist. */ int local_memory_node(int node) { struct zone *zone; (void)first_zones_zonelist(node_zonelist(node, GFP_KERNEL), gfp_zone(GFP_KERNEL), NULL, &zone); return zone->node; } #endif |
f0c0b2b80 change zonelist o... |
3403 |
|
1da177e4c Linux-2.6.12-rc2 |
3404 |
#else /* CONFIG_NUMA */ |
f0c0b2b80 change zonelist o... |
3405 3406 3407 3408 3409 3410 |
static void set_zonelist_order(void) { current_zonelist_order = ZONELIST_ORDER_ZONE; } static void build_zonelists(pg_data_t *pgdat) |
1da177e4c Linux-2.6.12-rc2 |
3411 |
{ |
19655d348 [PATCH] linearly ... |
3412 |
int node, local_node; |
54a6eb5c4 mm: use two zonel... |
3413 3414 |
enum zone_type j; struct zonelist *zonelist; |
1da177e4c Linux-2.6.12-rc2 |
3415 3416 |
local_node = pgdat->node_id; |
1da177e4c Linux-2.6.12-rc2 |
3417 |
|
54a6eb5c4 mm: use two zonel... |
3418 3419 |
zonelist = &pgdat->node_zonelists[0]; j = build_zonelists_node(pgdat, zonelist, 0, MAX_NR_ZONES - 1); |
1da177e4c Linux-2.6.12-rc2 |
3420 |
|
54a6eb5c4 mm: use two zonel... |
3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 |
/* * Now we build the zonelist so that it contains the zones * of all the other nodes. * We don't want to pressure a particular node, so when * building the zones for node N, we make sure that the * zones coming right after the local ones are those from * node N+1 (modulo N) */ for (node = local_node + 1; node < MAX_NUMNODES; node++) { if (!node_online(node)) continue; j = build_zonelists_node(NODE_DATA(node), zonelist, j, MAX_NR_ZONES - 1); |
1da177e4c Linux-2.6.12-rc2 |
3434 |
} |
54a6eb5c4 mm: use two zonel... |
3435 3436 3437 3438 3439 3440 |
for (node = 0; node < local_node; node++) { if (!node_online(node)) continue; j = build_zonelists_node(NODE_DATA(node), zonelist, j, MAX_NR_ZONES - 1); } |
dd1a239f6 mm: have zonelist... |
3441 3442 |
zonelist->_zonerefs[j].zone = NULL; zonelist->_zonerefs[j].zone_idx = 0; |
1da177e4c Linux-2.6.12-rc2 |
3443 |
} |
9276b1bc9 [PATCH] memory pa... |
3444 |
/* non-NUMA variant of zonelist performance cache - just NULL zlcache_ptr */ |
f0c0b2b80 change zonelist o... |
3445 |
static void build_zonelist_cache(pg_data_t *pgdat) |
9276b1bc9 [PATCH] memory pa... |
3446 |
{ |
54a6eb5c4 mm: use two zonel... |
3447 |
pgdat->node_zonelists[0].zlcache_ptr = NULL; |
9276b1bc9 [PATCH] memory pa... |
3448 |
} |
1da177e4c Linux-2.6.12-rc2 |
3449 |
#endif /* CONFIG_NUMA */ |
99dcc3e5a this_cpu: Page al... |
3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 |
/* * Boot pageset table. One per cpu which is going to be used for all * zones and all nodes. The parameters will be set in such a way * that an item put on a list will immediately be handed over to * the buddy list. This is safe since pageset manipulation is done * with interrupts disabled. * * The boot_pagesets must be kept even after bootup is complete for * unused processors and/or zones. They do play a role for bootstrapping * hotplugged processors. * * zoneinfo_show() and maybe other functions do * not check if the processor is online before following the pageset pointer. * Other parts of the kernel may not check if the zone is available. */ static void setup_pageset(struct per_cpu_pageset *p, unsigned long batch); static DEFINE_PER_CPU(struct per_cpu_pageset, boot_pageset); |
1f522509c mem-hotplug: avoi... |
3467 |
static void setup_zone_pageset(struct zone *zone); |
99dcc3e5a this_cpu: Page al... |
3468 |
|
4eaf3f643 mem-hotplug: fix ... |
3469 3470 3471 3472 3473 |
/* * Global mutex to protect against size modification of zonelists * as well as to serialize pageset setup for the new populated zone. */ DEFINE_MUTEX(zonelists_mutex); |
9b1a4d383 stop_machine: Wea... |
3474 |
/* return values int ....just for stop_machine() */ |
4ed7e0222 mm/hotplug: mark ... |
3475 |
static int __build_all_zonelists(void *data) |
1da177e4c Linux-2.6.12-rc2 |
3476 |
{ |
6811378e7 [PATCH] wait_tabl... |
3477 |
int nid; |
99dcc3e5a this_cpu: Page al... |
3478 |
int cpu; |
9adb62a5d mm/hotplug: corre... |
3479 |
pg_data_t *self = data; |
9276b1bc9 [PATCH] memory pa... |
3480 |
|
7f9cfb310 mm: build_zonelis... |
3481 3482 3483 |
#ifdef CONFIG_NUMA memset(node_load, 0, sizeof(node_load)); #endif |
9adb62a5d mm/hotplug: corre... |
3484 3485 3486 3487 3488 |
if (self && !node_online(self->node_id)) { build_zonelists(self); build_zonelist_cache(self); } |
9276b1bc9 [PATCH] memory pa... |
3489 |
for_each_online_node(nid) { |
7ea1530ab Memoryless nodes:... |
3490 3491 3492 3493 |
pg_data_t *pgdat = NODE_DATA(nid); build_zonelists(pgdat); build_zonelist_cache(pgdat); |
9276b1bc9 [PATCH] memory pa... |
3494 |
} |
99dcc3e5a this_cpu: Page al... |
3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 |
/* * Initialize the boot_pagesets that are going to be used * for bootstrapping processors. The real pagesets for * each zone will be allocated later when the per cpu * allocator is available. * * boot_pagesets are used also for bootstrapping offline * cpus if the system is already booted because the pagesets * are needed to initialize allocators on a specific cpu too. * F.e. the percpu allocator needs the page allocator which * needs the percpu allocator in order to allocate its pagesets * (a chicken-egg dilemma). */ |
7aac78988 numa: introduce n... |
3509 |
for_each_possible_cpu(cpu) { |
99dcc3e5a this_cpu: Page al... |
3510 |
setup_pageset(&per_cpu(boot_pageset, cpu), 0); |
7aac78988 numa: introduce n... |
3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 |
#ifdef CONFIG_HAVE_MEMORYLESS_NODES /* * We now know the "local memory node" for each node-- * i.e., the node of the first zone in the generic zonelist. * Set up numa_mem percpu variable for on-line cpus. During * boot, only the boot cpu should be on-line; we'll init the * secondary cpus' numa_mem as they come on-line. During * node/memory hotplug, we'll fixup all on-line cpus. */ if (cpu_online(cpu)) set_cpu_numa_mem(cpu, local_memory_node(cpu_to_node(cpu))); #endif } |
6811378e7 [PATCH] wait_tabl... |
3524 3525 |
return 0; } |
4eaf3f643 mem-hotplug: fix ... |
3526 3527 3528 3529 |
/* * Called with zonelists_mutex held always * unless system_state == SYSTEM_BOOTING. */ |
9adb62a5d mm/hotplug: corre... |
3530 |
void __ref build_all_zonelists(pg_data_t *pgdat, struct zone *zone) |
6811378e7 [PATCH] wait_tabl... |
3531 |
{ |
f0c0b2b80 change zonelist o... |
3532 |
set_zonelist_order(); |
6811378e7 [PATCH] wait_tabl... |
3533 |
if (system_state == SYSTEM_BOOTING) { |
423b41d77 [PATCH] mm/page_a... |
3534 |
__build_all_zonelists(NULL); |
68ad8df42 mm: print out the... |
3535 |
mminit_verify_zonelist(); |
6811378e7 [PATCH] wait_tabl... |
3536 3537 |
cpuset_init_current_mems_allowed(); } else { |
183ff22bb spelling fixes: mm/ |
3538 |
/* we have to stop all cpus to guarantee there is no user |
6811378e7 [PATCH] wait_tabl... |
3539 |
of zonelist */ |
e9959f0f3 mm/page_alloc.c: ... |
3540 |
#ifdef CONFIG_MEMORY_HOTPLUG |
9adb62a5d mm/hotplug: corre... |
3541 3542 |
if (zone) setup_zone_pageset(zone); |
e9959f0f3 mm/page_alloc.c: ... |
3543 |
#endif |
9adb62a5d mm/hotplug: corre... |
3544 |
stop_machine(__build_all_zonelists, pgdat, NULL); |
6811378e7 [PATCH] wait_tabl... |
3545 3546 |
/* cpuset refresh routine should be here */ } |
bd1e22b8e [PATCH] initialis... |
3547 |
vm_total_pages = nr_free_pagecache_pages(); |
9ef9acb05 Do not group page... |
3548 3549 3550 3551 3552 3553 3554 |
/* * Disable grouping by mobility if the number of pages in the * system is too low to allow the mechanism to work. It would be * more accurate, but expensive to check per-zone. This check is * made on memory-hotadd so a system can start with mobility * disabled and enable it later */ |
d9c234005 Do not depend on ... |
3555 |
if (vm_total_pages < (pageblock_nr_pages * MIGRATE_TYPES)) |
9ef9acb05 Do not group page... |
3556 3557 3558 3559 3560 3561 3562 |
page_group_by_mobility_disabled = 1; else page_group_by_mobility_disabled = 0; printk("Built %i zonelists in %s order, mobility grouping %s. " "Total pages: %ld ", |
62bc62a87 page allocator: u... |
3563 |
nr_online_nodes, |
f0c0b2b80 change zonelist o... |
3564 |
zonelist_order_name[current_zonelist_order], |
9ef9acb05 Do not group page... |
3565 |
page_group_by_mobility_disabled ? "off" : "on", |
f0c0b2b80 change zonelist o... |
3566 3567 3568 3569 3570 |
vm_total_pages); #ifdef CONFIG_NUMA printk("Policy zone: %s ", zone_names[policy_zone]); #endif |
1da177e4c Linux-2.6.12-rc2 |
3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 |
} /* * Helper functions to size the waitqueue hash table. * Essentially these want to choose hash table sizes sufficiently * large so that collisions trying to wait on pages are rare. * But in fact, the number of active page waitqueues on typical * systems is ridiculously low, less than 200. So this is even * conservative, even though it seems large. * * The constant PAGES_PER_WAITQUEUE specifies the ratio of pages to * waitqueues, i.e. the size of the waitq table given the number of pages. */ #define PAGES_PER_WAITQUEUE 256 |
cca448fe9 [PATCH] wait_tabl... |
3585 |
#ifndef CONFIG_MEMORY_HOTPLUG |
02b694dea [PATCH] wait_tabl... |
3586 |
static inline unsigned long wait_table_hash_nr_entries(unsigned long pages) |
1da177e4c Linux-2.6.12-rc2 |
3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 |
{ unsigned long size = 1; pages /= PAGES_PER_WAITQUEUE; while (size < pages) size <<= 1; /* * Once we have dozens or even hundreds of threads sleeping * on IO we've got bigger problems than wait queue collision. * Limit the size of the wait table to a reasonable size. */ size = min(size, 4096UL); return max(size, 4UL); } |
cca448fe9 [PATCH] wait_tabl... |
3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 |
#else /* * A zone's size might be changed by hot-add, so it is not possible to determine * a suitable size for its wait_table. So we use the maximum size now. * * The max wait table size = 4096 x sizeof(wait_queue_head_t). ie: * * i386 (preemption config) : 4096 x 16 = 64Kbyte. * ia64, x86-64 (no preemption): 4096 x 20 = 80Kbyte. * ia64, x86-64 (preemption) : 4096 x 24 = 96Kbyte. * * The maximum entries are prepared when a zone's memory is (512K + 256) pages * or more by the traditional way. (See above). It equals: * * i386, x86-64, powerpc(4K page size) : = ( 2G + 1M)byte. * ia64(16K page size) : = ( 8G + 4M)byte. * powerpc (64K page size) : = (32G +16M)byte. */ static inline unsigned long wait_table_hash_nr_entries(unsigned long pages) { return 4096UL; } #endif |
1da177e4c Linux-2.6.12-rc2 |
3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 |
/* * This is an integer logarithm so that shifts can be used later * to extract the more random high bits from the multiplicative * hash function before the remainder is taken. */ static inline unsigned long wait_table_bits(unsigned long size) { return ffz(~size); } #define LONG_ALIGN(x) (((x)+(sizeof(long))-1)&~((sizeof(long))-1)) |
56fd56b86 Bias the location... |
3639 |
/* |
6d3163ce8 mm: check if any ... |
3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 |
* Check if a pageblock contains reserved pages */ static int pageblock_is_reserved(unsigned long start_pfn, unsigned long end_pfn) { unsigned long pfn; for (pfn = start_pfn; pfn < end_pfn; pfn++) { if (!pfn_valid_within(pfn) || PageReserved(pfn_to_page(pfn))) return 1; } return 0; } /* |
d9c234005 Do not depend on ... |
3654 |
* Mark a number of pageblocks as MIGRATE_RESERVE. The number |
418589663 page allocator: u... |
3655 3656 |
* of blocks reserved is based on min_wmark_pages(zone). The memory within * the reserve will tend to store contiguous free pages. Setting min_free_kbytes |
56fd56b86 Bias the location... |
3657 3658 3659 3660 3661 |
* higher will lead to a bigger reserve which will get freed as contiguous * blocks as reclaim kicks in */ static void setup_zone_migrate_reserve(struct zone *zone) { |
6d3163ce8 mm: check if any ... |
3662 |
unsigned long start_pfn, pfn, end_pfn, block_end_pfn; |
56fd56b86 Bias the location... |
3663 |
struct page *page; |
78986a678 page-allocator: l... |
3664 3665 |
unsigned long block_migratetype; int reserve; |
56fd56b86 Bias the location... |
3666 |
|
d02156388 mm: Ensure that p... |
3667 3668 3669 3670 3671 3672 |
/* * Get the start pfn, end pfn and the number of blocks to reserve * We have to be careful to be aligned to pageblock_nr_pages to * make sure that we always check pfn_valid for the first page in * the block. */ |
56fd56b86 Bias the location... |
3673 |
start_pfn = zone->zone_start_pfn; |
108bcc96e mm: add & use zon... |
3674 |
end_pfn = zone_end_pfn(zone); |
d02156388 mm: Ensure that p... |
3675 |
start_pfn = roundup(start_pfn, pageblock_nr_pages); |
418589663 page allocator: u... |
3676 |
reserve = roundup(min_wmark_pages(zone), pageblock_nr_pages) >> |
d9c234005 Do not depend on ... |
3677 |
pageblock_order; |
56fd56b86 Bias the location... |
3678 |
|
78986a678 page-allocator: l... |
3679 3680 3681 3682 3683 3684 3685 3686 |
/* * Reserve blocks are generally in place to help high-order atomic * allocations that are short-lived. A min_free_kbytes value that * would result in more than 2 reserve blocks for atomic allocations * is assumed to be in place to help anti-fragmentation for the * future allocation of hugepages at runtime. */ reserve = min(2, reserve); |
d9c234005 Do not depend on ... |
3687 |
for (pfn = start_pfn; pfn < end_pfn; pfn += pageblock_nr_pages) { |
56fd56b86 Bias the location... |
3688 3689 3690 |
if (!pfn_valid(pfn)) continue; page = pfn_to_page(pfn); |
344c790e3 mm: make setup_zo... |
3691 3692 3693 |
/* Watch out for overlapping nodes */ if (page_to_nid(page) != zone_to_nid(zone)) continue; |
56fd56b86 Bias the location... |
3694 |
block_migratetype = get_pageblock_migratetype(page); |
938929f14 mm: reduce the am... |
3695 3696 3697 3698 3699 3700 3701 3702 3703 |
/* Only test what is necessary when the reserves are not met */ if (reserve > 0) { /* * Blocks with reserved pages will never free, skip * them. */ block_end_pfn = min(pfn + pageblock_nr_pages, end_pfn); if (pageblock_is_reserved(pfn, block_end_pfn)) continue; |
56fd56b86 Bias the location... |
3704 |
|
938929f14 mm: reduce the am... |
3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 |
/* If this block is reserved, account for it */ if (block_migratetype == MIGRATE_RESERVE) { reserve--; continue; } /* Suitable for reserving if this block is movable */ if (block_migratetype == MIGRATE_MOVABLE) { set_pageblock_migratetype(page, MIGRATE_RESERVE); move_freepages_block(zone, page, MIGRATE_RESERVE); reserve--; continue; } |
56fd56b86 Bias the location... |
3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 |
} /* * If the reserve is met and this is a previous reserved block, * take it back */ if (block_migratetype == MIGRATE_RESERVE) { set_pageblock_migratetype(page, MIGRATE_MOVABLE); move_freepages_block(zone, page, MIGRATE_MOVABLE); } } } |
ac0e5b7a6 remove PAGE_GROUP... |
3732 |
|
1da177e4c Linux-2.6.12-rc2 |
3733 3734 3735 3736 3737 |
/* * Initially all pages are reserved - free ones are freed * up by free_all_bootmem() once the early boot process is * done. Non-atomic initialization, single-pass. */ |
c09b42404 [PATCH] x86_64: a... |
3738 |
void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, |
a2f3aa025 [PATCH] Fix spars... |
3739 |
unsigned long start_pfn, enum memmap_context context) |
1da177e4c Linux-2.6.12-rc2 |
3740 |
{ |
1da177e4c Linux-2.6.12-rc2 |
3741 |
struct page *page; |
29751f699 [PATCH] sparsemem... |
3742 3743 |
unsigned long end_pfn = start_pfn + size; unsigned long pfn; |
86051ca5e mm: fix usemap in... |
3744 |
struct zone *z; |
1da177e4c Linux-2.6.12-rc2 |
3745 |
|
22b31eec6 badpage: vm_norma... |
3746 3747 |
if (highest_memmap_pfn < end_pfn - 1) highest_memmap_pfn = end_pfn - 1; |
86051ca5e mm: fix usemap in... |
3748 |
z = &NODE_DATA(nid)->node_zones[zone]; |
cbe8dd4af [PATCH] memmap_in... |
3749 |
for (pfn = start_pfn; pfn < end_pfn; pfn++) { |
a2f3aa025 [PATCH] Fix spars... |
3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 |
/* * There can be holes in boot-time mem_map[]s * handed to this function. They do not * exist on hotplugged memory. */ if (context == MEMMAP_EARLY) { if (!early_pfn_valid(pfn)) continue; if (!early_pfn_in_nid(pfn, nid)) continue; } |
d41dee369 [PATCH] sparsemem... |
3761 3762 |
page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); |
708614e61 mm: verify the pa... |
3763 |
mminit_verify_page_links(page, zone, nid, pfn); |
7835e98b2 [PATCH] remove se... |
3764 |
init_page_count(page); |
22b751c3d mm: rename page s... |
3765 3766 |
page_mapcount_reset(page); page_nid_reset_last(page); |
1da177e4c Linux-2.6.12-rc2 |
3767 |
SetPageReserved(page); |
b2a0ac887 Split the free li... |
3768 3769 3770 3771 3772 |
/* * Mark the block movable so that blocks are reserved for * movable at startup. This will force kernel allocations * to reserve their blocks rather than leaking throughout * the address space during boot when many long-lived |
56fd56b86 Bias the location... |
3773 3774 3775 |
* kernel allocations are made. Later some blocks near * the start are marked MIGRATE_RESERVE by * setup_zone_migrate_reserve() |
86051ca5e mm: fix usemap in... |
3776 3777 3778 3779 3780 |
* * bitmap is created for zone's valid pfn range. but memmap * can be created for invalid pages (for alignment) * check here not to call set_pageblock_migratetype() against * pfn out of zone. |
b2a0ac887 Split the free li... |
3781 |
*/ |
86051ca5e mm: fix usemap in... |
3782 |
if ((z->zone_start_pfn <= pfn) |
108bcc96e mm: add & use zon... |
3783 |
&& (pfn < zone_end_pfn(z)) |
86051ca5e mm: fix usemap in... |
3784 |
&& !(pfn & (pageblock_nr_pages - 1))) |
56fd56b86 Bias the location... |
3785 |
set_pageblock_migratetype(page, MIGRATE_MOVABLE); |
b2a0ac887 Split the free li... |
3786 |
|
1da177e4c Linux-2.6.12-rc2 |
3787 3788 3789 3790 |
INIT_LIST_HEAD(&page->lru); #ifdef WANT_PAGE_VIRTUAL /* The shift won't overflow because ZONE_NORMAL is below 4G. */ if (!is_highmem_idx(zone)) |
3212c6be2 [PATCH] fix WANT_... |
3791 |
set_page_address(page, __va(pfn << PAGE_SHIFT)); |
1da177e4c Linux-2.6.12-rc2 |
3792 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
3793 3794 |
} } |
1e548deb5 page allocator: r... |
3795 |
static void __meminit zone_init_free_lists(struct zone *zone) |
1da177e4c Linux-2.6.12-rc2 |
3796 |
{ |
b2a0ac887 Split the free li... |
3797 3798 3799 |
int order, t; for_each_migratetype_order(order, t) { INIT_LIST_HEAD(&zone->free_area[order].free_list[t]); |
1da177e4c Linux-2.6.12-rc2 |
3800 3801 3802 3803 3804 3805 |
zone->free_area[order].nr_free = 0; } } #ifndef __HAVE_ARCH_MEMMAP_INIT #define memmap_init(size, nid, zone, start_pfn) \ |
a2f3aa025 [PATCH] Fix spars... |
3806 |
memmap_init_zone((size), (nid), (zone), (start_pfn), MEMMAP_EARLY) |
1da177e4c Linux-2.6.12-rc2 |
3807 |
#endif |
4ed7e0222 mm/hotplug: mark ... |
3808 |
static int __meminit zone_batchsize(struct zone *zone) |
e7c8d5c99 [PATCH] node loca... |
3809 |
{ |
3a6be87fd nommu: clamp zone... |
3810 |
#ifdef CONFIG_MMU |
e7c8d5c99 [PATCH] node loca... |
3811 3812 3813 3814 |
int batch; /* * The per-cpu-pages pools are set to around 1000th of the |
ba56e91c9 [PATCH] mm: page_... |
3815 |
* size of the zone. But no more than 1/2 of a meg. |
e7c8d5c99 [PATCH] node loca... |
3816 3817 3818 |
* * OK, so we don't know how big the cache is. So guess. */ |
b40da0494 mm: use zone->pre... |
3819 |
batch = zone->managed_pages / 1024; |
ba56e91c9 [PATCH] mm: page_... |
3820 3821 |
if (batch * PAGE_SIZE > 512 * 1024) batch = (512 * 1024) / PAGE_SIZE; |
e7c8d5c99 [PATCH] node loca... |
3822 3823 3824 3825 3826 |
batch /= 4; /* We effectively *= 4 below */ if (batch < 1) batch = 1; /* |
0ceaacc97 [PATCH] Fix up pe... |
3827 3828 3829 |
* Clamp the batch to a 2^n - 1 value. Having a power * of 2 value was found to be more likely to have * suboptimal cache aliasing properties in some cases. |
e7c8d5c99 [PATCH] node loca... |
3830 |
* |
0ceaacc97 [PATCH] Fix up pe... |
3831 3832 3833 3834 |
* For example if 2 tasks are alternately allocating * batches of pages, one task can end up with a lot * of pages of one half of the possible page colors * and the other with pages of the other colors. |
e7c8d5c99 [PATCH] node loca... |
3835 |
*/ |
9155203a5 mm: use roundown_... |
3836 |
batch = rounddown_pow_of_two(batch + batch/2) - 1; |
ba56e91c9 [PATCH] mm: page_... |
3837 |
|
e7c8d5c99 [PATCH] node loca... |
3838 |
return batch; |
3a6be87fd nommu: clamp zone... |
3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 |
#else /* The deferral and batching of frees should be suppressed under NOMMU * conditions. * * The problem is that NOMMU needs to be able to allocate large chunks * of contiguous memory as there's no hardware page translation to * assemble apparent contiguous memory from discontiguous pages. * * Queueing large contiguous runs of pages for batching, however, * causes the pages to actually be freed in smaller chunks. As there * can be a significant delay between the individual batches being * recycled, this leads to the once large chunks of space being * fragmented and becoming unavailable for high-order allocations. */ return 0; #endif |
e7c8d5c99 [PATCH] node loca... |
3856 |
} |
b69a7288e mm/page_alloc.c: ... |
3857 |
static void setup_pageset(struct per_cpu_pageset *p, unsigned long batch) |
2caaad41e [PATCH] Reduce si... |
3858 3859 |
{ struct per_cpu_pages *pcp; |
5f8dcc212 page-allocator: s... |
3860 |
int migratetype; |
2caaad41e [PATCH] Reduce si... |
3861 |
|
1c6fe9465 [PATCH] NUMA: bro... |
3862 |
memset(p, 0, sizeof(*p)); |
3dfa5721f Page allocator: g... |
3863 |
pcp = &p->pcp; |
2caaad41e [PATCH] Reduce si... |
3864 |
pcp->count = 0; |
2caaad41e [PATCH] Reduce si... |
3865 3866 |
pcp->high = 6 * batch; pcp->batch = max(1UL, 1 * batch); |
5f8dcc212 page-allocator: s... |
3867 3868 |
for (migratetype = 0; migratetype < MIGRATE_PCPTYPES; migratetype++) INIT_LIST_HEAD(&pcp->lists[migratetype]); |
2caaad41e [PATCH] Reduce si... |
3869 |
} |
8ad4b1fb8 [PATCH] Make high... |
3870 3871 3872 3873 3874 3875 3876 3877 3878 |
/* * setup_pagelist_highmark() sets the high water mark for hot per_cpu_pagelist * to the value high for the pageset p. */ static void setup_pagelist_highmark(struct per_cpu_pageset *p, unsigned long high) { struct per_cpu_pages *pcp; |
3dfa5721f Page allocator: g... |
3879 |
pcp = &p->pcp; |
8ad4b1fb8 [PATCH] Make high... |
3880 3881 3882 3883 3884 |
pcp->high = high; pcp->batch = max(1UL, high/4); if ((high/4) > (PAGE_SHIFT * 8)) pcp->batch = PAGE_SHIFT * 8; } |
4ed7e0222 mm/hotplug: mark ... |
3885 |
static void __meminit setup_zone_pageset(struct zone *zone) |
319774e25 mem-hotplug: sepa... |
3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 |
{ int cpu; zone->pageset = alloc_percpu(struct per_cpu_pageset); for_each_possible_cpu(cpu) { struct per_cpu_pageset *pcp = per_cpu_ptr(zone->pageset, cpu); setup_pageset(pcp, zone_batchsize(zone)); if (percpu_pagelist_fraction) setup_pagelist_highmark(pcp, |
b40da0494 mm: use zone->pre... |
3898 |
(zone->managed_pages / |
319774e25 mem-hotplug: sepa... |
3899 3900 3901 |
percpu_pagelist_fraction)); } } |
2caaad41e [PATCH] Reduce si... |
3902 |
/* |
99dcc3e5a this_cpu: Page al... |
3903 3904 |
* Allocate per cpu pagesets and initialize them. * Before this call only boot pagesets were available. |
e7c8d5c99 [PATCH] node loca... |
3905 |
*/ |
99dcc3e5a this_cpu: Page al... |
3906 |
void __init setup_per_cpu_pageset(void) |
e7c8d5c99 [PATCH] node loca... |
3907 |
{ |
99dcc3e5a this_cpu: Page al... |
3908 |
struct zone *zone; |
e7c8d5c99 [PATCH] node loca... |
3909 |
|
319774e25 mem-hotplug: sepa... |
3910 3911 |
for_each_populated_zone(zone) setup_zone_pageset(zone); |
e7c8d5c99 [PATCH] node loca... |
3912 |
} |
577a32f62 mm: fix section m... |
3913 |
static noinline __init_refok |
cca448fe9 [PATCH] wait_tabl... |
3914 |
int zone_wait_table_init(struct zone *zone, unsigned long zone_size_pages) |
ed8ece2ec [PATCH] memory ho... |
3915 3916 3917 |
{ int i; struct pglist_data *pgdat = zone->zone_pgdat; |
cca448fe9 [PATCH] wait_tabl... |
3918 |
size_t alloc_size; |
ed8ece2ec [PATCH] memory ho... |
3919 3920 3921 3922 3923 |
/* * The per-page waitqueue mechanism uses hashed waitqueues * per zone. */ |
02b694dea [PATCH] wait_tabl... |
3924 3925 3926 3927 |
zone->wait_table_hash_nr_entries = wait_table_hash_nr_entries(zone_size_pages); zone->wait_table_bits = wait_table_bits(zone->wait_table_hash_nr_entries); |
cca448fe9 [PATCH] wait_tabl... |
3928 3929 |
alloc_size = zone->wait_table_hash_nr_entries * sizeof(wait_queue_head_t); |
cd94b9dbf memory hotplug: f... |
3930 |
if (!slab_is_available()) { |
cca448fe9 [PATCH] wait_tabl... |
3931 |
zone->wait_table = (wait_queue_head_t *) |
8f389a99b mm: use alloc_boo... |
3932 |
alloc_bootmem_node_nopanic(pgdat, alloc_size); |
cca448fe9 [PATCH] wait_tabl... |
3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 |
} else { /* * This case means that a zone whose size was 0 gets new memory * via memory hot-add. * But it may be the case that a new node was hot-added. In * this case vmalloc() will not be able to use this new node's * memory - this wait_table must be initialized to use this new * node itself as well. * To use this new node's memory, further consideration will be * necessary. */ |
8691f3a72 mm: no need to ca... |
3944 |
zone->wait_table = vmalloc(alloc_size); |
cca448fe9 [PATCH] wait_tabl... |
3945 3946 3947 |
} if (!zone->wait_table) return -ENOMEM; |
ed8ece2ec [PATCH] memory ho... |
3948 |
|
02b694dea [PATCH] wait_tabl... |
3949 |
for(i = 0; i < zone->wait_table_hash_nr_entries; ++i) |
ed8ece2ec [PATCH] memory ho... |
3950 |
init_waitqueue_head(zone->wait_table + i); |
cca448fe9 [PATCH] wait_tabl... |
3951 3952 |
return 0; |
ed8ece2ec [PATCH] memory ho... |
3953 |
} |
c09b42404 [PATCH] x86_64: a... |
3954 |
static __meminit void zone_pcp_init(struct zone *zone) |
ed8ece2ec [PATCH] memory ho... |
3955 |
{ |
99dcc3e5a this_cpu: Page al... |
3956 3957 3958 3959 3960 3961 |
/* * per cpu subsystem is not up at this point. The following code * relies on the ability of the linker to provide the * offset of a (static) per cpu variable into the per cpu area. */ zone->pageset = &boot_pageset; |
ed8ece2ec [PATCH] memory ho... |
3962 |
|
f5335c0f1 [PATCH] quieten z... |
3963 |
if (zone->present_pages) |
99dcc3e5a this_cpu: Page al... |
3964 3965 3966 3967 |
printk(KERN_DEBUG " %s zone: %lu pages, LIFO batch:%u ", zone->name, zone->present_pages, zone_batchsize(zone)); |
ed8ece2ec [PATCH] memory ho... |
3968 |
} |
4ed7e0222 mm/hotplug: mark ... |
3969 |
int __meminit init_currently_empty_zone(struct zone *zone, |
718127cc3 [PATCH] wait_tabl... |
3970 |
unsigned long zone_start_pfn, |
a2f3aa025 [PATCH] Fix spars... |
3971 3972 |
unsigned long size, enum memmap_context context) |
ed8ece2ec [PATCH] memory ho... |
3973 3974 |
{ struct pglist_data *pgdat = zone->zone_pgdat; |
cca448fe9 [PATCH] wait_tabl... |
3975 3976 3977 3978 |
int ret; ret = zone_wait_table_init(zone, size); if (ret) return ret; |
ed8ece2ec [PATCH] memory ho... |
3979 |
pgdat->nr_zones = zone_idx(zone) + 1; |
ed8ece2ec [PATCH] memory ho... |
3980 |
zone->zone_start_pfn = zone_start_pfn; |
708614e61 mm: verify the pa... |
3981 3982 3983 3984 3985 3986 |
mminit_dprintk(MMINIT_TRACE, "memmap_init", "Initialising map node %d zone %lu pfns %lu -> %lu ", pgdat->node_id, (unsigned long)zone_idx(zone), zone_start_pfn, (zone_start_pfn + size)); |
1e548deb5 page allocator: r... |
3987 |
zone_init_free_lists(zone); |
718127cc3 [PATCH] wait_tabl... |
3988 3989 |
return 0; |
ed8ece2ec [PATCH] memory ho... |
3990 |
} |
0ee332c14 memblock: Kill ea... |
3991 |
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP |
c713216de [PATCH] Introduce... |
3992 3993 3994 3995 3996 3997 3998 |
#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID /* * Required by SPARSEMEM. Given a PFN, return what node the PFN is on. * Architectures may implement their own version but if add_active_range() * was used and there are no special requirements, this is a convenient * alternative */ |
f2dbcfa73 mm: clean up for ... |
3999 |
int __meminit __early_pfn_to_nid(unsigned long pfn) |
c713216de [PATCH] Introduce... |
4000 |
{ |
c13291a53 bootmem: Use for_... |
4001 4002 |
unsigned long start_pfn, end_pfn; int i, nid; |
7c243c716 mm: speedup in __... |
4003 4004 4005 4006 4007 4008 4009 4010 4011 |
/* * NOTE: The following SMP-unsafe globals are only used early in boot * when the kernel is running single-threaded. */ static unsigned long __meminitdata last_start_pfn, last_end_pfn; static int __meminitdata last_nid; if (last_start_pfn <= pfn && pfn < last_end_pfn) return last_nid; |
c713216de [PATCH] Introduce... |
4012 |
|
c13291a53 bootmem: Use for_... |
4013 |
for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) |
7c243c716 mm: speedup in __... |
4014 4015 4016 4017 |
if (start_pfn <= pfn && pfn < end_pfn) { last_start_pfn = start_pfn; last_end_pfn = end_pfn; last_nid = nid; |
c13291a53 bootmem: Use for_... |
4018 |
return nid; |
7c243c716 mm: speedup in __... |
4019 |
} |
cc2559bcc mm: fix memmap in... |
4020 4021 |
/* This is a memory hole */ return -1; |
c713216de [PATCH] Introduce... |
4022 4023 |
} #endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ |
f2dbcfa73 mm: clean up for ... |
4024 4025 |
int __meminit early_pfn_to_nid(unsigned long pfn) { |
cc2559bcc mm: fix memmap in... |
4026 4027 4028 4029 4030 4031 4032 |
int nid; nid = __early_pfn_to_nid(pfn); if (nid >= 0) return nid; /* just returns 0 */ return 0; |
f2dbcfa73 mm: clean up for ... |
4033 |
} |
cc2559bcc mm: fix memmap in... |
4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 |
#ifdef CONFIG_NODES_SPAN_OTHER_NODES bool __meminit early_pfn_in_nid(unsigned long pfn, int node) { int nid; nid = __early_pfn_to_nid(pfn); if (nid >= 0 && nid != node) return false; return true; } #endif |
f2dbcfa73 mm: clean up for ... |
4045 |
|
c713216de [PATCH] Introduce... |
4046 4047 |
/** * free_bootmem_with_active_regions - Call free_bootmem_node for each active range |
88ca3b94e [PATCH] page_allo... |
4048 4049 |
* @nid: The node to free memory on. If MAX_NUMNODES, all nodes are freed. * @max_low_pfn: The highest PFN that will be passed to free_bootmem_node |
c713216de [PATCH] Introduce... |
4050 4051 4052 4053 4054 |
* * If an architecture guarantees that all ranges registered with * add_active_ranges() contain no holes and may be freed, this * this function may be used instead of calling free_bootmem() manually. */ |
c13291a53 bootmem: Use for_... |
4055 |
void __init free_bootmem_with_active_regions(int nid, unsigned long max_low_pfn) |
cc2898943 mm: Move early_no... |
4056 |
{ |
c13291a53 bootmem: Use for_... |
4057 4058 |
unsigned long start_pfn, end_pfn; int i, this_nid; |
edbe7d23b memblock: Add fin... |
4059 |
|
c13291a53 bootmem: Use for_... |
4060 4061 4062 |
for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, &this_nid) { start_pfn = min(start_pfn, max_low_pfn); end_pfn = min(end_pfn, max_low_pfn); |
edbe7d23b memblock: Add fin... |
4063 |
|
c13291a53 bootmem: Use for_... |
4064 4065 4066 4067 |
if (start_pfn < end_pfn) free_bootmem_node(NODE_DATA(this_nid), PFN_PHYS(start_pfn), (end_pfn - start_pfn) << PAGE_SHIFT); |
edbe7d23b memblock: Add fin... |
4068 |
} |
edbe7d23b memblock: Add fin... |
4069 |
} |
edbe7d23b memblock: Add fin... |
4070 |
|
c713216de [PATCH] Introduce... |
4071 4072 |
/** * sparse_memory_present_with_active_regions - Call memory_present for each active range |
88ca3b94e [PATCH] page_allo... |
4073 |
* @nid: The node to call memory_present for. If MAX_NUMNODES, all nodes will be used. |
c713216de [PATCH] Introduce... |
4074 4075 4076 |
* * If an architecture guarantees that all ranges registered with * add_active_ranges() contain no holes and may be freed, this |
88ca3b94e [PATCH] page_allo... |
4077 |
* function may be used instead of calling memory_present() manually. |
c713216de [PATCH] Introduce... |
4078 4079 4080 |
*/ void __init sparse_memory_present_with_active_regions(int nid) { |
c13291a53 bootmem: Use for_... |
4081 4082 |
unsigned long start_pfn, end_pfn; int i, this_nid; |
c713216de [PATCH] Introduce... |
4083 |
|
c13291a53 bootmem: Use for_... |
4084 4085 |
for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, &this_nid) memory_present(this_nid, start_pfn, end_pfn); |
c713216de [PATCH] Introduce... |
4086 4087 4088 4089 |
} /** * get_pfn_range_for_nid - Return the start and end page frames for a node |
88ca3b94e [PATCH] page_allo... |
4090 4091 4092 |
* @nid: The nid to return the range for. If MAX_NUMNODES, the min and max PFN are returned. * @start_pfn: Passed by reference. On return, it will have the node start_pfn. * @end_pfn: Passed by reference. On return, it will have the node end_pfn. |
c713216de [PATCH] Introduce... |
4093 4094 4095 4096 |
* * It returns the start and end page frame of a node based on information * provided by an arch calling add_active_range(). If called for a node * with no available memory, a warning is printed and the start and end |
88ca3b94e [PATCH] page_allo... |
4097 |
* PFNs will be 0. |
c713216de [PATCH] Introduce... |
4098 |
*/ |
a3142c8e1 Fix section misma... |
4099 |
void __meminit get_pfn_range_for_nid(unsigned int nid, |
c713216de [PATCH] Introduce... |
4100 4101 |
unsigned long *start_pfn, unsigned long *end_pfn) { |
c13291a53 bootmem: Use for_... |
4102 |
unsigned long this_start_pfn, this_end_pfn; |
c713216de [PATCH] Introduce... |
4103 |
int i; |
c13291a53 bootmem: Use for_... |
4104 |
|
c713216de [PATCH] Introduce... |
4105 4106 |
*start_pfn = -1UL; *end_pfn = 0; |
c13291a53 bootmem: Use for_... |
4107 4108 4109 |
for_each_mem_pfn_range(i, nid, &this_start_pfn, &this_end_pfn, NULL) { *start_pfn = min(*start_pfn, this_start_pfn); *end_pfn = max(*end_pfn, this_end_pfn); |
c713216de [PATCH] Introduce... |
4110 |
} |
633c0666b Memoryless nodes:... |
4111 |
if (*start_pfn == -1UL) |
c713216de [PATCH] Introduce... |
4112 |
*start_pfn = 0; |
c713216de [PATCH] Introduce... |
4113 4114 4115 |
} /* |
2a1e274ac Create the ZONE_M... |
4116 4117 4118 4119 |
* This finds a zone that can be used for ZONE_MOVABLE pages. The * assumption is made that zones within a node are ordered in monotonic * increasing memory addresses so that the "highest" populated zone is used */ |
b69a7288e mm/page_alloc.c: ... |
4120 |
static void __init find_usable_zone_for_movable(void) |
2a1e274ac Create the ZONE_M... |
4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 |
{ int zone_index; for (zone_index = MAX_NR_ZONES - 1; zone_index >= 0; zone_index--) { if (zone_index == ZONE_MOVABLE) continue; if (arch_zone_highest_possible_pfn[zone_index] > arch_zone_lowest_possible_pfn[zone_index]) break; } VM_BUG_ON(zone_index == -1); movable_zone = zone_index; } /* * The zone ranges provided by the architecture do not include ZONE_MOVABLE |
25985edce Fix common misspe... |
4138 |
* because it is sized independent of architecture. Unlike the other zones, |
2a1e274ac Create the ZONE_M... |
4139 4140 4141 4142 4143 4144 4145 |
* the starting point for ZONE_MOVABLE is not fixed. It may be different * in each node depending on the size of each node and how evenly kernelcore * is distributed. This helper function adjusts the zone ranges * provided by the architecture for a given node by using the end of the * highest usable zone for ZONE_MOVABLE. This preserves the assumption that * zones within a node are in order of monotonic increases memory addresses */ |
b69a7288e mm/page_alloc.c: ... |
4146 |
static void __meminit adjust_zone_range_for_zone_movable(int nid, |
2a1e274ac Create the ZONE_M... |
4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 |
unsigned long zone_type, unsigned long node_start_pfn, unsigned long node_end_pfn, unsigned long *zone_start_pfn, unsigned long *zone_end_pfn) { /* Only adjust if ZONE_MOVABLE is on this node */ if (zone_movable_pfn[nid]) { /* Size ZONE_MOVABLE */ if (zone_type == ZONE_MOVABLE) { *zone_start_pfn = zone_movable_pfn[nid]; *zone_end_pfn = min(node_end_pfn, arch_zone_highest_possible_pfn[movable_zone]); /* Adjust for ZONE_MOVABLE starting within this range */ } else if (*zone_start_pfn < zone_movable_pfn[nid] && *zone_end_pfn > zone_movable_pfn[nid]) { *zone_end_pfn = zone_movable_pfn[nid]; /* Check if this whole range is within ZONE_MOVABLE */ } else if (*zone_start_pfn >= zone_movable_pfn[nid]) *zone_start_pfn = *zone_end_pfn; } } /* |
c713216de [PATCH] Introduce... |
4173 4174 4175 |
* Return the number of pages a zone spans in a node, including holes * present_pages = zone_spanned_pages_in_node() - zone_absent_pages_in_node() */ |
6ea6e6887 mm: more __memini... |
4176 |
static unsigned long __meminit zone_spanned_pages_in_node(int nid, |
c713216de [PATCH] Introduce... |
4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 |
unsigned long zone_type, unsigned long *ignored) { unsigned long node_start_pfn, node_end_pfn; unsigned long zone_start_pfn, zone_end_pfn; /* Get the start and end of the node and zone */ get_pfn_range_for_nid(nid, &node_start_pfn, &node_end_pfn); zone_start_pfn = arch_zone_lowest_possible_pfn[zone_type]; zone_end_pfn = arch_zone_highest_possible_pfn[zone_type]; |
2a1e274ac Create the ZONE_M... |
4187 4188 4189 |
adjust_zone_range_for_zone_movable(nid, zone_type, node_start_pfn, node_end_pfn, &zone_start_pfn, &zone_end_pfn); |
c713216de [PATCH] Introduce... |
4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 |
/* Check that this node has pages within the zone's required range */ if (zone_end_pfn < node_start_pfn || zone_start_pfn > node_end_pfn) return 0; /* Move the zone boundaries inside the node if necessary */ zone_end_pfn = min(zone_end_pfn, node_end_pfn); zone_start_pfn = max(zone_start_pfn, node_start_pfn); /* Return the spanned pages */ return zone_end_pfn - zone_start_pfn; } /* * Return the number of holes in a range on a node. If nid is MAX_NUMNODES, |
88ca3b94e [PATCH] page_allo... |
4205 |
* then all holes in the requested range will be accounted for. |
c713216de [PATCH] Introduce... |
4206 |
*/ |
329962503 x86: Fix checking... |
4207 |
unsigned long __meminit __absent_pages_in_range(int nid, |
c713216de [PATCH] Introduce... |
4208 4209 4210 |
unsigned long range_start_pfn, unsigned long range_end_pfn) { |
96e907d13 bootmem: Reimplem... |
4211 4212 4213 |
unsigned long nr_absent = range_end_pfn - range_start_pfn; unsigned long start_pfn, end_pfn; int i; |
c713216de [PATCH] Introduce... |
4214 |
|
96e907d13 bootmem: Reimplem... |
4215 4216 4217 4218 |
for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, NULL) { start_pfn = clamp(start_pfn, range_start_pfn, range_end_pfn); end_pfn = clamp(end_pfn, range_start_pfn, range_end_pfn); nr_absent -= end_pfn - start_pfn; |
c713216de [PATCH] Introduce... |
4219 |
} |
96e907d13 bootmem: Reimplem... |
4220 |
return nr_absent; |
c713216de [PATCH] Introduce... |
4221 4222 4223 4224 4225 4226 4227 |
} /** * absent_pages_in_range - Return number of page frames in holes within a range * @start_pfn: The start PFN to start searching for holes * @end_pfn: The end PFN to stop searching for holes * |
88ca3b94e [PATCH] page_allo... |
4228 |
* It returns the number of pages frames in memory holes within a range. |
c713216de [PATCH] Introduce... |
4229 4230 4231 4232 4233 4234 4235 4236 |
*/ unsigned long __init absent_pages_in_range(unsigned long start_pfn, unsigned long end_pfn) { return __absent_pages_in_range(MAX_NUMNODES, start_pfn, end_pfn); } /* Return the number of page frames in holes in a zone on a node */ |
6ea6e6887 mm: more __memini... |
4237 |
static unsigned long __meminit zone_absent_pages_in_node(int nid, |
c713216de [PATCH] Introduce... |
4238 4239 4240 |
unsigned long zone_type, unsigned long *ignored) { |
96e907d13 bootmem: Reimplem... |
4241 4242 |
unsigned long zone_low = arch_zone_lowest_possible_pfn[zone_type]; unsigned long zone_high = arch_zone_highest_possible_pfn[zone_type]; |
9c7cd6877 [PATCH] Account f... |
4243 4244 4245 4246 |
unsigned long node_start_pfn, node_end_pfn; unsigned long zone_start_pfn, zone_end_pfn; get_pfn_range_for_nid(nid, &node_start_pfn, &node_end_pfn); |
96e907d13 bootmem: Reimplem... |
4247 4248 |
zone_start_pfn = clamp(node_start_pfn, zone_low, zone_high); zone_end_pfn = clamp(node_end_pfn, zone_low, zone_high); |
9c7cd6877 [PATCH] Account f... |
4249 |
|
2a1e274ac Create the ZONE_M... |
4250 4251 4252 |
adjust_zone_range_for_zone_movable(nid, zone_type, node_start_pfn, node_end_pfn, &zone_start_pfn, &zone_end_pfn); |
9c7cd6877 [PATCH] Account f... |
4253 |
return __absent_pages_in_range(nid, zone_start_pfn, zone_end_pfn); |
c713216de [PATCH] Introduce... |
4254 |
} |
0e0b864e0 [PATCH] Account f... |
4255 |
|
0ee332c14 memblock: Kill ea... |
4256 |
#else /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ |
6ea6e6887 mm: more __memini... |
4257 |
static inline unsigned long __meminit zone_spanned_pages_in_node(int nid, |
c713216de [PATCH] Introduce... |
4258 4259 4260 4261 4262 |
unsigned long zone_type, unsigned long *zones_size) { return zones_size[zone_type]; } |
6ea6e6887 mm: more __memini... |
4263 |
static inline unsigned long __meminit zone_absent_pages_in_node(int nid, |
c713216de [PATCH] Introduce... |
4264 4265 4266 4267 4268 4269 4270 4271 |
unsigned long zone_type, unsigned long *zholes_size) { if (!zholes_size) return 0; return zholes_size[zone_type]; } |
20e6926dc x86, ACPI, mm: Re... |
4272 |
|
0ee332c14 memblock: Kill ea... |
4273 |
#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ |
c713216de [PATCH] Introduce... |
4274 |
|
a3142c8e1 Fix section misma... |
4275 |
static void __meminit calculate_node_totalpages(struct pglist_data *pgdat, |
c713216de [PATCH] Introduce... |
4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 |
unsigned long *zones_size, unsigned long *zholes_size) { unsigned long realtotalpages, totalpages = 0; enum zone_type i; for (i = 0; i < MAX_NR_ZONES; i++) totalpages += zone_spanned_pages_in_node(pgdat->node_id, i, zones_size); pgdat->node_spanned_pages = totalpages; realtotalpages = totalpages; for (i = 0; i < MAX_NR_ZONES; i++) realtotalpages -= zone_absent_pages_in_node(pgdat->node_id, i, zholes_size); pgdat->node_present_pages = realtotalpages; printk(KERN_DEBUG "On node %d totalpages: %lu ", pgdat->node_id, realtotalpages); } |
835c134ec Add a bitmap that... |
4296 4297 4298 |
#ifndef CONFIG_SPARSEMEM /* * Calculate the size of the zone->blockflags rounded to an unsigned long |
d9c234005 Do not depend on ... |
4299 4300 |
* Start by making sure zonesize is a multiple of pageblock_order by rounding * up. Then use 1 NR_PAGEBLOCK_BITS worth of bits per pageblock, finally |
835c134ec Add a bitmap that... |
4301 4302 4303 |
* round what is now in bits to nearest long in bits, then return it in * bytes. */ |
7c45512df mm: fix pageblock... |
4304 |
static unsigned long __init usemap_size(unsigned long zone_start_pfn, unsigned long zonesize) |
835c134ec Add a bitmap that... |
4305 4306 |
{ unsigned long usemapsize; |
7c45512df mm: fix pageblock... |
4307 |
zonesize += zone_start_pfn & (pageblock_nr_pages-1); |
d9c234005 Do not depend on ... |
4308 4309 |
usemapsize = roundup(zonesize, pageblock_nr_pages); usemapsize = usemapsize >> pageblock_order; |
835c134ec Add a bitmap that... |
4310 4311 4312 4313 4314 4315 4316 |
usemapsize *= NR_PAGEBLOCK_BITS; usemapsize = roundup(usemapsize, 8 * sizeof(unsigned long)); return usemapsize / 8; } static void __init setup_usemap(struct pglist_data *pgdat, |
7c45512df mm: fix pageblock... |
4317 4318 4319 |
struct zone *zone, unsigned long zone_start_pfn, unsigned long zonesize) |
835c134ec Add a bitmap that... |
4320 |
{ |
7c45512df mm: fix pageblock... |
4321 |
unsigned long usemapsize = usemap_size(zone_start_pfn, zonesize); |
835c134ec Add a bitmap that... |
4322 |
zone->pageblock_flags = NULL; |
58a01a457 mm/page_alloc.c: ... |
4323 |
if (usemapsize) |
8f389a99b mm: use alloc_boo... |
4324 4325 |
zone->pageblock_flags = alloc_bootmem_node_nopanic(pgdat, usemapsize); |
835c134ec Add a bitmap that... |
4326 4327 |
} #else |
7c45512df mm: fix pageblock... |
4328 4329 |
static inline void setup_usemap(struct pglist_data *pgdat, struct zone *zone, unsigned long zone_start_pfn, unsigned long zonesize) {} |
835c134ec Add a bitmap that... |
4330 |
#endif /* CONFIG_SPARSEMEM */ |
d9c234005 Do not depend on ... |
4331 |
#ifdef CONFIG_HUGETLB_PAGE_SIZE_VARIABLE |
ba72cb8cb Fix boot problem ... |
4332 |
|
d9c234005 Do not depend on ... |
4333 |
/* Initialise the number of pages represented by NR_PAGEBLOCK_BITS */ |
ca57df79d mm: setup pageblo... |
4334 |
void __init set_pageblock_order(void) |
d9c234005 Do not depend on ... |
4335 |
{ |
955c1cd74 mm/page_alloc.c: ... |
4336 |
unsigned int order; |
d9c234005 Do not depend on ... |
4337 4338 4339 |
/* Check that pageblock_nr_pages has not already been setup */ if (pageblock_order) return; |
955c1cd74 mm/page_alloc.c: ... |
4340 4341 4342 4343 |
if (HPAGE_SHIFT > PAGE_SHIFT) order = HUGETLB_PAGE_ORDER; else order = MAX_ORDER - 1; |
d9c234005 Do not depend on ... |
4344 4345 |
/* * Assume the largest contiguous order of interest is a huge page. |
955c1cd74 mm/page_alloc.c: ... |
4346 4347 |
* This value may be variable depending on boot parameters on IA64 and * powerpc. |
d9c234005 Do not depend on ... |
4348 4349 4350 4351 |
*/ pageblock_order = order; } #else /* CONFIG_HUGETLB_PAGE_SIZE_VARIABLE */ |
ba72cb8cb Fix boot problem ... |
4352 4353 |
/* * When CONFIG_HUGETLB_PAGE_SIZE_VARIABLE is not set, set_pageblock_order() |
955c1cd74 mm/page_alloc.c: ... |
4354 4355 4356 |
* is unused as pageblock_order is set at compile-time. See * include/linux/pageblock-flags.h for the values of pageblock_order based on * the kernel config |
ba72cb8cb Fix boot problem ... |
4357 |
*/ |
ca57df79d mm: setup pageblo... |
4358 |
void __init set_pageblock_order(void) |
ba72cb8cb Fix boot problem ... |
4359 |
{ |
ba72cb8cb Fix boot problem ... |
4360 |
} |
d9c234005 Do not depend on ... |
4361 4362 |
#endif /* CONFIG_HUGETLB_PAGE_SIZE_VARIABLE */ |
01cefaef4 mm: provide more ... |
4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 |
static unsigned long __paginginit calc_memmap_size(unsigned long spanned_pages, unsigned long present_pages) { unsigned long pages = spanned_pages; /* * Provide a more accurate estimation if there are holes within * the zone and SPARSEMEM is in use. If there are holes within the * zone, each populated memory region may cost us one or two extra * memmap pages due to alignment because memmap pages for each * populated regions may not naturally algined on page boundary. * So the (present_pages >> 4) heuristic is a tradeoff for that. */ if (spanned_pages > present_pages + (present_pages >> 4) && IS_ENABLED(CONFIG_SPARSEMEM)) pages = present_pages; return PAGE_ALIGN(pages * sizeof(struct page)) >> PAGE_SHIFT; } |
1da177e4c Linux-2.6.12-rc2 |
4382 4383 4384 4385 4386 |
/* * Set up the zone data structures: * - mark all pages reserved * - mark all memory queues empty * - clear the memory bitmaps |
6527af5d1 mm: remove redund... |
4387 4388 |
* * NOTE: pgdat should get zeroed by caller. |
1da177e4c Linux-2.6.12-rc2 |
4389 |
*/ |
b5a0e0113 Solve section mis... |
4390 |
static void __paginginit free_area_init_core(struct pglist_data *pgdat, |
1da177e4c Linux-2.6.12-rc2 |
4391 4392 |
unsigned long *zones_size, unsigned long *zholes_size) { |
2f1b62486 [PATCH] reduce MA... |
4393 |
enum zone_type j; |
ed8ece2ec [PATCH] memory ho... |
4394 |
int nid = pgdat->node_id; |
1da177e4c Linux-2.6.12-rc2 |
4395 |
unsigned long zone_start_pfn = pgdat->node_start_pfn; |
718127cc3 [PATCH] wait_tabl... |
4396 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
4397 |
|
208d54e55 [PATCH] memory ho... |
4398 |
pgdat_resize_init(pgdat); |
8177a420e mm: numa: Structu... |
4399 4400 4401 4402 4403 |
#ifdef CONFIG_NUMA_BALANCING spin_lock_init(&pgdat->numabalancing_migrate_lock); pgdat->numabalancing_migrate_nr_pages = 0; pgdat->numabalancing_migrate_next_window = jiffies; #endif |
1da177e4c Linux-2.6.12-rc2 |
4404 |
init_waitqueue_head(&pgdat->kswapd_wait); |
5515061d2 mm: throttle dire... |
4405 |
init_waitqueue_head(&pgdat->pfmemalloc_wait); |
52d4b9ac0 memcg: allocate a... |
4406 |
pgdat_page_cgroup_init(pgdat); |
5f63b720b mm: page_alloc: r... |
4407 |
|
1da177e4c Linux-2.6.12-rc2 |
4408 4409 |
for (j = 0; j < MAX_NR_ZONES; j++) { struct zone *zone = pgdat->node_zones + j; |
9feedc9d8 mm: introduce new... |
4410 |
unsigned long size, realsize, freesize, memmap_pages; |
1da177e4c Linux-2.6.12-rc2 |
4411 |
|
c713216de [PATCH] Introduce... |
4412 |
size = zone_spanned_pages_in_node(nid, j, zones_size); |
9feedc9d8 mm: introduce new... |
4413 |
realsize = freesize = size - zone_absent_pages_in_node(nid, j, |
c713216de [PATCH] Introduce... |
4414 |
zholes_size); |
1da177e4c Linux-2.6.12-rc2 |
4415 |
|
0e0b864e0 [PATCH] Account f... |
4416 |
/* |
9feedc9d8 mm: introduce new... |
4417 |
* Adjust freesize so that it accounts for how much memory |
0e0b864e0 [PATCH] Account f... |
4418 4419 4420 |
* is used by this zone for memmap. This affects the watermark * and per-cpu initialisations */ |
01cefaef4 mm: provide more ... |
4421 |
memmap_pages = calc_memmap_size(size, realsize); |
9feedc9d8 mm: introduce new... |
4422 4423 |
if (freesize >= memmap_pages) { freesize -= memmap_pages; |
5594c8c81 mm: print out mem... |
4424 4425 4426 4427 4428 |
if (memmap_pages) printk(KERN_DEBUG " %s zone: %lu pages used for memmap ", zone_names[j], memmap_pages); |
0e0b864e0 [PATCH] Account f... |
4429 4430 |
} else printk(KERN_WARNING |
9feedc9d8 mm: introduce new... |
4431 4432 4433 |
" %s zone: %lu pages exceeds freesize %lu ", zone_names[j], memmap_pages, freesize); |
0e0b864e0 [PATCH] Account f... |
4434 |
|
6267276f3 [PATCH] optional ... |
4435 |
/* Account for reserved pages */ |
9feedc9d8 mm: introduce new... |
4436 4437 |
if (j == 0 && freesize > dma_reserve) { freesize -= dma_reserve; |
d903ef9f3 mm: print out mem... |
4438 4439 |
printk(KERN_DEBUG " %s zone: %lu pages reserved ", |
6267276f3 [PATCH] optional ... |
4440 |
zone_names[0], dma_reserve); |
0e0b864e0 [PATCH] Account f... |
4441 |
} |
98d2b0ebd [PATCH] reduce MA... |
4442 |
if (!is_highmem_idx(j)) |
9feedc9d8 mm: introduce new... |
4443 |
nr_kernel_pages += freesize; |
01cefaef4 mm: provide more ... |
4444 4445 4446 |
/* Charge for highmem memmap if there are enough kernel pages */ else if (nr_kernel_pages > memmap_pages * 2) nr_kernel_pages -= memmap_pages; |
9feedc9d8 mm: introduce new... |
4447 |
nr_all_pages += freesize; |
1da177e4c Linux-2.6.12-rc2 |
4448 4449 |
zone->spanned_pages = size; |
306f2e9ee mm: set zone->pre... |
4450 |
zone->present_pages = realsize; |
9feedc9d8 mm: introduce new... |
4451 4452 4453 4454 4455 4456 |
/* * Set an approximate value for lowmem here, it will be adjusted * when the bootmem allocator frees pages into the buddy system. * And all highmem pages will be managed by the buddy system. */ zone->managed_pages = is_highmem_idx(j) ? realsize : freesize; |
9614634fe [PATCH] ZVC/zone_... |
4457 |
#ifdef CONFIG_NUMA |
d5f541ed6 [PATCH] Add node ... |
4458 |
zone->node = nid; |
9feedc9d8 mm: introduce new... |
4459 |
zone->min_unmapped_pages = (freesize*sysctl_min_unmapped_ratio) |
9614634fe [PATCH] ZVC/zone_... |
4460 |
/ 100; |
9feedc9d8 mm: introduce new... |
4461 |
zone->min_slab_pages = (freesize * sysctl_min_slab_ratio) / 100; |
9614634fe [PATCH] ZVC/zone_... |
4462 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
4463 4464 4465 |
zone->name = zone_names[j]; spin_lock_init(&zone->lock); spin_lock_init(&zone->lru_lock); |
bdc8cb984 [PATCH] memory ho... |
4466 |
zone_seqlock_init(zone); |
1da177e4c Linux-2.6.12-rc2 |
4467 |
zone->zone_pgdat = pgdat; |
1da177e4c Linux-2.6.12-rc2 |
4468 |
|
ed8ece2ec [PATCH] memory ho... |
4469 |
zone_pcp_init(zone); |
bea8c150a memcg: fix hotplu... |
4470 |
lruvec_init(&zone->lruvec); |
1da177e4c Linux-2.6.12-rc2 |
4471 4472 |
if (!size) continue; |
955c1cd74 mm/page_alloc.c: ... |
4473 |
set_pageblock_order(); |
7c45512df mm: fix pageblock... |
4474 |
setup_usemap(pgdat, zone, zone_start_pfn, size); |
a2f3aa025 [PATCH] Fix spars... |
4475 4476 |
ret = init_currently_empty_zone(zone, zone_start_pfn, size, MEMMAP_EARLY); |
718127cc3 [PATCH] wait_tabl... |
4477 |
BUG_ON(ret); |
76cdd58e5 memory_hotplug: a... |
4478 |
memmap_init(size, nid, j, zone_start_pfn); |
1da177e4c Linux-2.6.12-rc2 |
4479 |
zone_start_pfn += size; |
1da177e4c Linux-2.6.12-rc2 |
4480 4481 |
} } |
577a32f62 mm: fix section m... |
4482 |
static void __init_refok alloc_node_mem_map(struct pglist_data *pgdat) |
1da177e4c Linux-2.6.12-rc2 |
4483 |
{ |
1da177e4c Linux-2.6.12-rc2 |
4484 4485 4486 |
/* Skip empty nodes */ if (!pgdat->node_spanned_pages) return; |
d41dee369 [PATCH] sparsemem... |
4487 |
#ifdef CONFIG_FLAT_NODE_MEM_MAP |
1da177e4c Linux-2.6.12-rc2 |
4488 4489 |
/* ia64 gets its own node_mem_map, before this, without bootmem */ if (!pgdat->node_mem_map) { |
e984bb43f [PATCH] Align the... |
4490 |
unsigned long size, start, end; |
d41dee369 [PATCH] sparsemem... |
4491 |
struct page *map; |
e984bb43f [PATCH] Align the... |
4492 4493 4494 4495 4496 4497 |
/* * The zone's endpoints aren't required to be MAX_ORDER * aligned but the node_mem_map endpoints must be in order * for the buddy allocator to function correctly. */ start = pgdat->node_start_pfn & ~(MAX_ORDER_NR_PAGES - 1); |
108bcc96e mm: add & use zon... |
4498 |
end = pgdat_end_pfn(pgdat); |
e984bb43f [PATCH] Align the... |
4499 4500 |
end = ALIGN(end, MAX_ORDER_NR_PAGES); size = (end - start) * sizeof(struct page); |
6f167ec72 [PATCH] sparsemem... |
4501 4502 |
map = alloc_remap(pgdat->node_id, size); if (!map) |
8f389a99b mm: use alloc_boo... |
4503 |
map = alloc_bootmem_node_nopanic(pgdat, size); |
e984bb43f [PATCH] Align the... |
4504 |
pgdat->node_mem_map = map + (pgdat->node_start_pfn - start); |
1da177e4c Linux-2.6.12-rc2 |
4505 |
} |
12d810c1b m68k: discontinuo... |
4506 |
#ifndef CONFIG_NEED_MULTIPLE_NODES |
1da177e4c Linux-2.6.12-rc2 |
4507 4508 4509 |
/* * With no DISCONTIG, the global mem_map is just set as node 0's */ |
c713216de [PATCH] Introduce... |
4510 |
if (pgdat == NODE_DATA(0)) { |
1da177e4c Linux-2.6.12-rc2 |
4511 |
mem_map = NODE_DATA(0)->node_mem_map; |
0ee332c14 memblock: Kill ea... |
4512 |
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP |
c713216de [PATCH] Introduce... |
4513 |
if (page_to_pfn(mem_map) != pgdat->node_start_pfn) |
467bc461d Fix crash with FL... |
4514 |
mem_map -= (pgdat->node_start_pfn - ARCH_PFN_OFFSET); |
0ee332c14 memblock: Kill ea... |
4515 |
#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ |
c713216de [PATCH] Introduce... |
4516 |
} |
1da177e4c Linux-2.6.12-rc2 |
4517 |
#endif |
d41dee369 [PATCH] sparsemem... |
4518 |
#endif /* CONFIG_FLAT_NODE_MEM_MAP */ |
1da177e4c Linux-2.6.12-rc2 |
4519 |
} |
9109fb7b3 mm: drop unneeded... |
4520 4521 |
void __paginginit free_area_init_node(int nid, unsigned long *zones_size, unsigned long node_start_pfn, unsigned long *zholes_size) |
1da177e4c Linux-2.6.12-rc2 |
4522 |
{ |
9109fb7b3 mm: drop unneeded... |
4523 |
pg_data_t *pgdat = NODE_DATA(nid); |
88fdf75d1 mm: warn if pg_da... |
4524 |
/* pg_data_t should be reset to zero when it's allocated */ |
8783b6e2b mm: remove node_s... |
4525 |
WARN_ON(pgdat->nr_zones || pgdat->classzone_idx); |
88fdf75d1 mm: warn if pg_da... |
4526 |
|
1da177e4c Linux-2.6.12-rc2 |
4527 4528 |
pgdat->node_id = nid; pgdat->node_start_pfn = node_start_pfn; |
957f822a0 mm, numa: reclaim... |
4529 |
init_zone_allows_reclaim(nid); |
c713216de [PATCH] Introduce... |
4530 |
calculate_node_totalpages(pgdat, zones_size, zholes_size); |
1da177e4c Linux-2.6.12-rc2 |
4531 4532 |
alloc_node_mem_map(pgdat); |
e8c27ac91 x86, numa, 32-bit... |
4533 4534 4535 4536 4537 4538 |
#ifdef CONFIG_FLAT_NODE_MEM_MAP printk(KERN_DEBUG "free_area_init_node: node %d, pgdat %08lx, node_mem_map %08lx ", nid, (unsigned long)pgdat, (unsigned long)pgdat->node_mem_map); #endif |
1da177e4c Linux-2.6.12-rc2 |
4539 4540 4541 |
free_area_init_core(pgdat, zones_size, zholes_size); } |
0ee332c14 memblock: Kill ea... |
4542 |
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP |
418508c13 fix unused setup_... |
4543 4544 4545 4546 4547 |
#if MAX_NUMNODES > 1 /* * Figure out the number of possible node ids. */ |
f9872caf0 page_alloc: make ... |
4548 |
void __init setup_nr_node_ids(void) |
418508c13 fix unused setup_... |
4549 4550 4551 4552 4553 4554 4555 4556 |
{ unsigned int node; unsigned int highest = 0; for_each_node_mask(node, node_possible_map) highest = node; nr_node_ids = highest + 1; } |
418508c13 fix unused setup_... |
4557 |
#endif |
c713216de [PATCH] Introduce... |
4558 |
/** |
1e01979c8 x86, numa: Implem... |
4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 |
* node_map_pfn_alignment - determine the maximum internode alignment * * This function should be called after node map is populated and sorted. * It calculates the maximum power of two alignment which can distinguish * all the nodes. * * For example, if all nodes are 1GiB and aligned to 1GiB, the return value * would indicate 1GiB alignment with (1 << (30 - PAGE_SHIFT)). If the * nodes are shifted by 256MiB, 256MiB. Note that if only the last node is * shifted, 1GiB is enough and this function will indicate so. * * This is used to test whether pfn -> nid mapping of the chosen memory * model has fine enough granularity to avoid incorrect mapping for the * populated node map. * * Returns the determined alignment in pfn's. 0 if there is no alignment * requirement (single node). */ unsigned long __init node_map_pfn_alignment(void) { unsigned long accl_mask = 0, last_end = 0; |
c13291a53 bootmem: Use for_... |
4580 |
unsigned long start, end, mask; |
1e01979c8 x86, numa: Implem... |
4581 |
int last_nid = -1; |
c13291a53 bootmem: Use for_... |
4582 |
int i, nid; |
1e01979c8 x86, numa: Implem... |
4583 |
|
c13291a53 bootmem: Use for_... |
4584 |
for_each_mem_pfn_range(i, MAX_NUMNODES, &start, &end, &nid) { |
1e01979c8 x86, numa: Implem... |
4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 |
if (!start || last_nid < 0 || last_nid == nid) { last_nid = nid; last_end = end; continue; } /* * Start with a mask granular enough to pin-point to the * start pfn and tick off bits one-by-one until it becomes * too coarse to separate the current node from the last. */ mask = ~((1 << __ffs(start)) - 1); while (mask && last_end <= (start & (mask << 1))) mask <<= 1; /* accumulate all internode masks */ accl_mask |= mask; } /* convert mask to number of pages */ return ~accl_mask + 1; } |
a6af2bc3d [PATCH] Avoid exc... |
4607 |
/* Find the lowest pfn for a node */ |
b69a7288e mm/page_alloc.c: ... |
4608 |
static unsigned long __init find_min_pfn_for_node(int nid) |
c713216de [PATCH] Introduce... |
4609 |
{ |
a6af2bc3d [PATCH] Avoid exc... |
4610 |
unsigned long min_pfn = ULONG_MAX; |
c13291a53 bootmem: Use for_... |
4611 4612 |
unsigned long start_pfn; int i; |
1abbfb412 [PATCH] x86_64: f... |
4613 |
|
c13291a53 bootmem: Use for_... |
4614 4615 |
for_each_mem_pfn_range(i, nid, &start_pfn, NULL, NULL) min_pfn = min(min_pfn, start_pfn); |
c713216de [PATCH] Introduce... |
4616 |
|
a6af2bc3d [PATCH] Avoid exc... |
4617 4618 |
if (min_pfn == ULONG_MAX) { printk(KERN_WARNING |
2bc0d2615 x86 boot: more co... |
4619 4620 |
"Could not find start_pfn for node %d ", nid); |
a6af2bc3d [PATCH] Avoid exc... |
4621 4622 4623 4624 |
return 0; } return min_pfn; |
c713216de [PATCH] Introduce... |
4625 4626 4627 4628 4629 4630 |
} /** * find_min_pfn_with_active_regions - Find the minimum PFN registered * * It returns the minimum PFN based on information provided via |
88ca3b94e [PATCH] page_allo... |
4631 |
* add_active_range(). |
c713216de [PATCH] Introduce... |
4632 4633 4634 4635 4636 |
*/ unsigned long __init find_min_pfn_with_active_regions(void) { return find_min_pfn_for_node(MAX_NUMNODES); } |
37b07e416 memoryless nodes:... |
4637 4638 4639 |
/* * early_calculate_totalpages() * Sum pages in active regions for movable zone. |
4b0ef1fe8 page_alloc: use N... |
4640 |
* Populate N_MEMORY for calculating usable_nodes. |
37b07e416 memoryless nodes:... |
4641 |
*/ |
484f51f82 mm/page_alloc.c: ... |
4642 |
static unsigned long __init early_calculate_totalpages(void) |
7e63efef8 Add a movablecore... |
4643 |
{ |
7e63efef8 Add a movablecore... |
4644 |
unsigned long totalpages = 0; |
c13291a53 bootmem: Use for_... |
4645 4646 4647 4648 4649 |
unsigned long start_pfn, end_pfn; int i, nid; for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) { unsigned long pages = end_pfn - start_pfn; |
7e63efef8 Add a movablecore... |
4650 |
|
37b07e416 memoryless nodes:... |
4651 4652 |
totalpages += pages; if (pages) |
4b0ef1fe8 page_alloc: use N... |
4653 |
node_set_state(nid, N_MEMORY); |
37b07e416 memoryless nodes:... |
4654 4655 |
} return totalpages; |
7e63efef8 Add a movablecore... |
4656 |
} |
2a1e274ac Create the ZONE_M... |
4657 4658 4659 4660 4661 4662 |
/* * Find the PFN the Movable zone begins in each node. Kernel memory * is spread evenly between nodes as long as the nodes have enough * memory. When they don't, some nodes will have more kernelcore than * others */ |
b224ef856 page_alloc: remov... |
4663 |
static void __init find_zone_movable_pfns_for_nodes(void) |
2a1e274ac Create the ZONE_M... |
4664 4665 4666 4667 |
{ int i, nid; unsigned long usable_startpfn; unsigned long kernelcore_node, kernelcore_remaining; |
66918dcdf x86: only clear n... |
4668 |
/* save the state before borrow the nodemask */ |
4b0ef1fe8 page_alloc: use N... |
4669 |
nodemask_t saved_node_state = node_states[N_MEMORY]; |
37b07e416 memoryless nodes:... |
4670 |
unsigned long totalpages = early_calculate_totalpages(); |
4b0ef1fe8 page_alloc: use N... |
4671 |
int usable_nodes = nodes_weight(node_states[N_MEMORY]); |
2a1e274ac Create the ZONE_M... |
4672 |
|
7e63efef8 Add a movablecore... |
4673 4674 4675 4676 4677 4678 4679 4680 4681 |
/* * If movablecore was specified, calculate what size of * kernelcore that corresponds so that memory usable for * any allocation type is evenly spread. If both kernelcore * and movablecore are specified, then the value of kernelcore * will be used for required_kernelcore if it's greater than * what movablecore would have allowed. */ if (required_movablecore) { |
7e63efef8 Add a movablecore... |
4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 |
unsigned long corepages; /* * Round-up so that ZONE_MOVABLE is at least as large as what * was requested by the user */ required_movablecore = roundup(required_movablecore, MAX_ORDER_NR_PAGES); corepages = totalpages - required_movablecore; required_kernelcore = max(required_kernelcore, corepages); } |
20e6926dc x86, ACPI, mm: Re... |
4694 4695 |
/* If kernelcore was not specified, there is no ZONE_MOVABLE */ if (!required_kernelcore) |
66918dcdf x86: only clear n... |
4696 |
goto out; |
2a1e274ac Create the ZONE_M... |
4697 4698 |
/* usable_startpfn is the lowest possible pfn ZONE_MOVABLE can be at */ |
20e6926dc x86, ACPI, mm: Re... |
4699 |
find_usable_zone_for_movable(); |
2a1e274ac Create the ZONE_M... |
4700 4701 4702 4703 4704 |
usable_startpfn = arch_zone_lowest_possible_pfn[movable_zone]; restart: /* Spread kernelcore memory as evenly as possible throughout nodes */ kernelcore_node = required_kernelcore / usable_nodes; |
4b0ef1fe8 page_alloc: use N... |
4705 |
for_each_node_state(nid, N_MEMORY) { |
c13291a53 bootmem: Use for_... |
4706 |
unsigned long start_pfn, end_pfn; |
2a1e274ac Create the ZONE_M... |
4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 |
/* * Recalculate kernelcore_node if the division per node * now exceeds what is necessary to satisfy the requested * amount of memory for the kernel */ if (required_kernelcore < kernelcore_node) kernelcore_node = required_kernelcore / usable_nodes; /* * As the map is walked, we track how much memory is usable * by the kernel using kernelcore_remaining. When it is * 0, the rest of the node is usable by ZONE_MOVABLE */ kernelcore_remaining = kernelcore_node; /* Go through each range of PFNs within this node */ |
c13291a53 bootmem: Use for_... |
4723 |
for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, NULL) { |
2a1e274ac Create the ZONE_M... |
4724 |
unsigned long size_pages; |
c13291a53 bootmem: Use for_... |
4725 |
start_pfn = max(start_pfn, zone_movable_pfn[nid]); |
2a1e274ac Create the ZONE_M... |
4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 |
if (start_pfn >= end_pfn) continue; /* Account for what is only usable for kernelcore */ if (start_pfn < usable_startpfn) { unsigned long kernel_pages; kernel_pages = min(end_pfn, usable_startpfn) - start_pfn; kernelcore_remaining -= min(kernel_pages, kernelcore_remaining); required_kernelcore -= min(kernel_pages, required_kernelcore); /* Continue if range is now fully accounted */ if (end_pfn <= usable_startpfn) { /* * Push zone_movable_pfn to the end so * that if we have to rebalance * kernelcore across nodes, we will * not double account here */ zone_movable_pfn[nid] = end_pfn; continue; } start_pfn = usable_startpfn; } /* * The usable PFN range for ZONE_MOVABLE is from * start_pfn->end_pfn. Calculate size_pages as the * number of pages used as kernelcore */ size_pages = end_pfn - start_pfn; if (size_pages > kernelcore_remaining) size_pages = kernelcore_remaining; zone_movable_pfn[nid] = start_pfn + size_pages; /* * Some kernelcore has been met, update counts and * break if the kernelcore for this node has been * satisified */ required_kernelcore -= min(required_kernelcore, size_pages); kernelcore_remaining -= size_pages; if (!kernelcore_remaining) break; } } /* * If there is still required_kernelcore, we do another pass with one * less node in the count. This will push zone_movable_pfn[nid] further * along on the nodes that still have memory until kernelcore is * satisified */ usable_nodes--; if (usable_nodes && required_kernelcore > usable_nodes) goto restart; /* Align start of ZONE_MOVABLE on all nids to MAX_ORDER_NR_PAGES */ for (nid = 0; nid < MAX_NUMNODES; nid++) zone_movable_pfn[nid] = roundup(zone_movable_pfn[nid], MAX_ORDER_NR_PAGES); |
66918dcdf x86: only clear n... |
4792 |
|
20e6926dc x86, ACPI, mm: Re... |
4793 |
out: |
66918dcdf x86: only clear n... |
4794 |
/* restore the node_state */ |
4b0ef1fe8 page_alloc: use N... |
4795 |
node_states[N_MEMORY] = saved_node_state; |
2a1e274ac Create the ZONE_M... |
4796 |
} |
4b0ef1fe8 page_alloc: use N... |
4797 4798 |
/* Any regular or high memory on that node ? */ static void check_for_memory(pg_data_t *pgdat, int nid) |
37b07e416 memoryless nodes:... |
4799 |
{ |
37b07e416 memoryless nodes:... |
4800 |
enum zone_type zone_type; |
4b0ef1fe8 page_alloc: use N... |
4801 4802 4803 4804 |
if (N_MEMORY == N_NORMAL_MEMORY) return; for (zone_type = 0; zone_type <= ZONE_MOVABLE - 1; zone_type++) { |
37b07e416 memoryless nodes:... |
4805 |
struct zone *zone = &pgdat->node_zones[zone_type]; |
d0048b0e5 page_alloc: break... |
4806 |
if (zone->present_pages) { |
4b0ef1fe8 page_alloc: use N... |
4807 4808 4809 4810 |
node_set_state(nid, N_HIGH_MEMORY); if (N_NORMAL_MEMORY != N_HIGH_MEMORY && zone_type <= ZONE_NORMAL) node_set_state(nid, N_NORMAL_MEMORY); |
d0048b0e5 page_alloc: break... |
4811 4812 |
break; } |
37b07e416 memoryless nodes:... |
4813 |
} |
37b07e416 memoryless nodes:... |
4814 |
} |
c713216de [PATCH] Introduce... |
4815 4816 |
/** * free_area_init_nodes - Initialise all pg_data_t and zone data |
88ca3b94e [PATCH] page_allo... |
4817 |
* @max_zone_pfn: an array of max PFNs for each zone |
c713216de [PATCH] Introduce... |
4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 |
* * This will call free_area_init_node() for each active node in the system. * Using the page ranges provided by add_active_range(), the size of each * zone in each node and their holes is calculated. If the maximum PFN * between two adjacent zones match, it is assumed that the zone is empty. * For example, if arch_max_dma_pfn == arch_max_dma32_pfn, it is assumed * that arch_max_dma32_pfn has no pages. It is also assumed that a zone * starts where the previous one ended. For example, ZONE_DMA32 starts * at arch_max_dma_pfn. */ void __init free_area_init_nodes(unsigned long *max_zone_pfn) { |
c13291a53 bootmem: Use for_... |
4830 4831 |
unsigned long start_pfn, end_pfn; int i, nid; |
a6af2bc3d [PATCH] Avoid exc... |
4832 |
|
c713216de [PATCH] Introduce... |
4833 4834 4835 4836 4837 4838 4839 4840 |
/* Record where the zone boundaries are */ memset(arch_zone_lowest_possible_pfn, 0, sizeof(arch_zone_lowest_possible_pfn)); memset(arch_zone_highest_possible_pfn, 0, sizeof(arch_zone_highest_possible_pfn)); arch_zone_lowest_possible_pfn[0] = find_min_pfn_with_active_regions(); arch_zone_highest_possible_pfn[0] = max_zone_pfn[0]; for (i = 1; i < MAX_NR_ZONES; i++) { |
2a1e274ac Create the ZONE_M... |
4841 4842 |
if (i == ZONE_MOVABLE) continue; |
c713216de [PATCH] Introduce... |
4843 4844 4845 4846 4847 |
arch_zone_lowest_possible_pfn[i] = arch_zone_highest_possible_pfn[i-1]; arch_zone_highest_possible_pfn[i] = max(max_zone_pfn[i], arch_zone_lowest_possible_pfn[i]); } |
2a1e274ac Create the ZONE_M... |
4848 4849 4850 4851 4852 |
arch_zone_lowest_possible_pfn[ZONE_MOVABLE] = 0; arch_zone_highest_possible_pfn[ZONE_MOVABLE] = 0; /* Find the PFNs that ZONE_MOVABLE begins at in each node */ memset(zone_movable_pfn, 0, sizeof(zone_movable_pfn)); |
b224ef856 page_alloc: remov... |
4853 |
find_zone_movable_pfns_for_nodes(); |
c713216de [PATCH] Introduce... |
4854 |
|
c713216de [PATCH] Introduce... |
4855 |
/* Print out the zone ranges */ |
a62e2f4f5 mm: print physica... |
4856 4857 |
printk("Zone ranges: "); |
2a1e274ac Create the ZONE_M... |
4858 4859 4860 |
for (i = 0; i < MAX_NR_ZONES; i++) { if (i == ZONE_MOVABLE) continue; |
155cbfc80 mm: use KERN_CONT... |
4861 |
printk(KERN_CONT " %-8s ", zone_names[i]); |
72f0ba025 mm: suppress pfn ... |
4862 4863 |
if (arch_zone_lowest_possible_pfn[i] == arch_zone_highest_possible_pfn[i]) |
155cbfc80 mm: use KERN_CONT... |
4864 4865 |
printk(KERN_CONT "empty "); |
72f0ba025 mm: suppress pfn ... |
4866 |
else |
a62e2f4f5 mm: print physica... |
4867 4868 4869 4870 4871 |
printk(KERN_CONT "[mem %0#10lx-%0#10lx] ", arch_zone_lowest_possible_pfn[i] << PAGE_SHIFT, (arch_zone_highest_possible_pfn[i] << PAGE_SHIFT) - 1); |
2a1e274ac Create the ZONE_M... |
4872 4873 4874 |
} /* Print out the PFNs ZONE_MOVABLE begins at in each node */ |
a62e2f4f5 mm: print physica... |
4875 4876 |
printk("Movable zone start for each node "); |
2a1e274ac Create the ZONE_M... |
4877 4878 |
for (i = 0; i < MAX_NUMNODES; i++) { if (zone_movable_pfn[i]) |
a62e2f4f5 mm: print physica... |
4879 4880 4881 |
printk(" Node %d: %#010lx ", i, zone_movable_pfn[i] << PAGE_SHIFT); |
2a1e274ac Create the ZONE_M... |
4882 |
} |
c713216de [PATCH] Introduce... |
4883 |
|
f2d52fe51 mm/memblock: clea... |
4884 |
/* Print out the early node map */ |
a62e2f4f5 mm: print physica... |
4885 4886 |
printk("Early memory node ranges "); |
c13291a53 bootmem: Use for_... |
4887 |
for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) |
a62e2f4f5 mm: print physica... |
4888 4889 4890 |
printk(" node %3d: [mem %#010lx-%#010lx] ", nid, start_pfn << PAGE_SHIFT, (end_pfn << PAGE_SHIFT) - 1); |
c713216de [PATCH] Introduce... |
4891 4892 |
/* Initialise every node */ |
708614e61 mm: verify the pa... |
4893 |
mminit_verify_pageflags_layout(); |
8ef828668 [PATCH] slab: red... |
4894 |
setup_nr_node_ids(); |
c713216de [PATCH] Introduce... |
4895 4896 |
for_each_online_node(nid) { pg_data_t *pgdat = NODE_DATA(nid); |
9109fb7b3 mm: drop unneeded... |
4897 |
free_area_init_node(nid, NULL, |
c713216de [PATCH] Introduce... |
4898 |
find_min_pfn_for_node(nid), NULL); |
37b07e416 memoryless nodes:... |
4899 4900 4901 |
/* Any memory on that node */ if (pgdat->node_present_pages) |
4b0ef1fe8 page_alloc: use N... |
4902 4903 |
node_set_state(nid, N_MEMORY); check_for_memory(pgdat, nid); |
c713216de [PATCH] Introduce... |
4904 4905 |
} } |
2a1e274ac Create the ZONE_M... |
4906 |
|
7e63efef8 Add a movablecore... |
4907 |
static int __init cmdline_parse_core(char *p, unsigned long *core) |
2a1e274ac Create the ZONE_M... |
4908 4909 4910 4911 4912 4913 |
{ unsigned long long coremem; if (!p) return -EINVAL; coremem = memparse(p, &p); |
7e63efef8 Add a movablecore... |
4914 |
*core = coremem >> PAGE_SHIFT; |
2a1e274ac Create the ZONE_M... |
4915 |
|
7e63efef8 Add a movablecore... |
4916 |
/* Paranoid check that UL is enough for the coremem value */ |
2a1e274ac Create the ZONE_M... |
4917 4918 4919 4920 |
WARN_ON((coremem >> PAGE_SHIFT) > ULONG_MAX); return 0; } |
ed7ed3651 handle kernelcore... |
4921 |
|
7e63efef8 Add a movablecore... |
4922 4923 4924 4925 4926 4927 4928 4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 |
/* * kernelcore=size sets the amount of memory for use for allocations that * cannot be reclaimed or migrated. */ static int __init cmdline_parse_kernelcore(char *p) { return cmdline_parse_core(p, &required_kernelcore); } /* * movablecore=size sets the amount of memory for use for allocations that * can be reclaimed or migrated. */ static int __init cmdline_parse_movablecore(char *p) { return cmdline_parse_core(p, &required_movablecore); } |
ed7ed3651 handle kernelcore... |
4939 |
early_param("kernelcore", cmdline_parse_kernelcore); |
7e63efef8 Add a movablecore... |
4940 |
early_param("movablecore", cmdline_parse_movablecore); |
ed7ed3651 handle kernelcore... |
4941 |
|
0ee332c14 memblock: Kill ea... |
4942 |
#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ |
c713216de [PATCH] Introduce... |
4943 |
|
69afade72 mm: introduce com... |
4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 |
unsigned long free_reserved_area(unsigned long start, unsigned long end, int poison, char *s) { unsigned long pages, pos; pos = start = PAGE_ALIGN(start); end &= PAGE_MASK; for (pages = 0; pos < end; pos += PAGE_SIZE, pages++) { if (poison) memset((void *)pos, poison, PAGE_SIZE); |
bb3ec6b08 mm: Fix virt_to_p... |
4954 |
free_reserved_page(virt_to_page((void *)pos)); |
69afade72 mm: introduce com... |
4955 4956 4957 4958 4959 4960 4961 4962 4963 |
} if (pages && s) pr_info("Freeing %s memory: %ldK (%lx - %lx) ", s, pages << (PAGE_SHIFT - 10), start, end); return pages; } |
cfa11e08e mm: introduce fre... |
4964 4965 4966 4967 4968 4969 4970 4971 |
#ifdef CONFIG_HIGHMEM void free_highmem_page(struct page *page) { __free_reserved_page(page); totalram_pages++; totalhigh_pages++; } #endif |
0e0b864e0 [PATCH] Account f... |
4972 |
/** |
88ca3b94e [PATCH] page_allo... |
4973 4974 |
* set_dma_reserve - set the specified number of pages reserved in the first zone * @new_dma_reserve: The number of pages to mark reserved |
0e0b864e0 [PATCH] Account f... |
4975 4976 4977 4978 |
* * The per-cpu batchsize and zone watermarks are determined by present_pages. * In the DMA zone, a significant percentage may be consumed by kernel image * and other unfreeable allocations which can skew the watermarks badly. This |
88ca3b94e [PATCH] page_allo... |
4979 4980 4981 |
* function may optionally be used to account for unfreeable pages in the * first zone (e.g., ZONE_DMA). The effect will be lower watermarks and * smaller per-cpu batchsize. |
0e0b864e0 [PATCH] Account f... |
4982 4983 4984 4985 4986 |
*/ void __init set_dma_reserve(unsigned long new_dma_reserve) { dma_reserve = new_dma_reserve; } |
1da177e4c Linux-2.6.12-rc2 |
4987 4988 |
void __init free_area_init(unsigned long *zones_size) { |
9109fb7b3 mm: drop unneeded... |
4989 |
free_area_init_node(0, zones_size, |
1da177e4c Linux-2.6.12-rc2 |
4990 4991 |
__pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL); } |
1da177e4c Linux-2.6.12-rc2 |
4992 |
|
1da177e4c Linux-2.6.12-rc2 |
4993 4994 4995 4996 |
static int page_alloc_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { int cpu = (unsigned long)hcpu; |
1da177e4c Linux-2.6.12-rc2 |
4997 |
|
8bb784428 Add suspend-relat... |
4998 |
if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) { |
f0cb3c76a mm: drain percpu ... |
4999 |
lru_add_drain_cpu(cpu); |
9f8f21725 Page allocator: c... |
5000 5001 5002 5003 5004 5005 5006 5007 |
drain_pages(cpu); /* * Spill the event counters of the dead processor * into the current processors event counters. * This artificially elevates the count of the current * processor. */ |
f8891e5e1 [PATCH] Light wei... |
5008 |
vm_events_fold_cpu(cpu); |
9f8f21725 Page allocator: c... |
5009 5010 5011 5012 5013 5014 5015 5016 |
/* * Zero the differential counters of the dead processor * so that the vm statistics are consistent. * * This is only okay since the processor is dead and cannot * race with what we are doing. */ |
2244b95a7 [PATCH] zoned vm ... |
5017 |
refresh_cpu_vm_stats(cpu); |
1da177e4c Linux-2.6.12-rc2 |
5018 5019 5020 |
} return NOTIFY_OK; } |
1da177e4c Linux-2.6.12-rc2 |
5021 5022 5023 5024 5025 5026 5027 |
void __init page_alloc_init(void) { hotcpu_notifier(page_alloc_cpu_notify, 0); } /* |
cb45b0e96 [PATCH] overcommi... |
5028 5029 5030 5031 5032 5033 5034 |
* calculate_totalreserve_pages - called when sysctl_lower_zone_reserve_ratio * or min_free_kbytes changes. */ static void calculate_totalreserve_pages(void) { struct pglist_data *pgdat; unsigned long reserve_pages = 0; |
2f6726e54 [PATCH] Apply typ... |
5035 |
enum zone_type i, j; |
cb45b0e96 [PATCH] overcommi... |
5036 5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 |
for_each_online_pgdat(pgdat) { for (i = 0; i < MAX_NR_ZONES; i++) { struct zone *zone = pgdat->node_zones + i; unsigned long max = 0; /* Find valid and maximum lowmem_reserve in the zone */ for (j = i; j < MAX_NR_ZONES; j++) { if (zone->lowmem_reserve[j] > max) max = zone->lowmem_reserve[j]; } |
418589663 page allocator: u... |
5047 5048 |
/* we treat the high watermark as reserved pages. */ max += high_wmark_pages(zone); |
cb45b0e96 [PATCH] overcommi... |
5049 |
|
b40da0494 mm: use zone->pre... |
5050 5051 |
if (max > zone->managed_pages) max = zone->managed_pages; |
cb45b0e96 [PATCH] overcommi... |
5052 |
reserve_pages += max; |
ab8fabd46 mm: exclude reser... |
5053 5054 5055 5056 5057 5058 5059 5060 5061 5062 |
/* * Lowmem reserves are not available to * GFP_HIGHUSER page cache allocations and * kswapd tries to balance zones to their high * watermark. As a result, neither should be * regarded as dirtyable memory, to prevent a * situation where reclaim has to clean pages * in order to balance the zones. */ zone->dirty_balance_reserve = max; |
cb45b0e96 [PATCH] overcommi... |
5063 5064 |
} } |
ab8fabd46 mm: exclude reser... |
5065 |
dirty_balance_reserve = reserve_pages; |
cb45b0e96 [PATCH] overcommi... |
5066 5067 5068 5069 |
totalreserve_pages = reserve_pages; } /* |
1da177e4c Linux-2.6.12-rc2 |
5070 5071 5072 5073 5074 5075 5076 5077 |
* setup_per_zone_lowmem_reserve - called whenever * sysctl_lower_zone_reserve_ratio changes. Ensures that each zone * has a correct pages reserved value, so an adequate number of * pages are left in the zone after a successful __alloc_pages(). */ static void setup_per_zone_lowmem_reserve(void) { struct pglist_data *pgdat; |
2f6726e54 [PATCH] Apply typ... |
5078 |
enum zone_type j, idx; |
1da177e4c Linux-2.6.12-rc2 |
5079 |
|
ec936fc56 [PATCH] for_each_... |
5080 |
for_each_online_pgdat(pgdat) { |
1da177e4c Linux-2.6.12-rc2 |
5081 5082 |
for (j = 0; j < MAX_NR_ZONES; j++) { struct zone *zone = pgdat->node_zones + j; |
b40da0494 mm: use zone->pre... |
5083 |
unsigned long managed_pages = zone->managed_pages; |
1da177e4c Linux-2.6.12-rc2 |
5084 5085 |
zone->lowmem_reserve[j] = 0; |
2f6726e54 [PATCH] Apply typ... |
5086 5087 |
idx = j; while (idx) { |
1da177e4c Linux-2.6.12-rc2 |
5088 |
struct zone *lower_zone; |
2f6726e54 [PATCH] Apply typ... |
5089 |
idx--; |
1da177e4c Linux-2.6.12-rc2 |
5090 5091 5092 5093 |
if (sysctl_lowmem_reserve_ratio[idx] < 1) sysctl_lowmem_reserve_ratio[idx] = 1; lower_zone = pgdat->node_zones + idx; |
b40da0494 mm: use zone->pre... |
5094 |
lower_zone->lowmem_reserve[j] = managed_pages / |
1da177e4c Linux-2.6.12-rc2 |
5095 |
sysctl_lowmem_reserve_ratio[idx]; |
b40da0494 mm: use zone->pre... |
5096 |
managed_pages += lower_zone->managed_pages; |
1da177e4c Linux-2.6.12-rc2 |
5097 5098 5099 |
} } } |
cb45b0e96 [PATCH] overcommi... |
5100 5101 5102 |
/* update totalreserve_pages */ calculate_totalreserve_pages(); |
1da177e4c Linux-2.6.12-rc2 |
5103 |
} |
cfd3da1e4 mm: Serialize acc... |
5104 |
static void __setup_per_zone_wmarks(void) |
1da177e4c Linux-2.6.12-rc2 |
5105 5106 5107 5108 5109 5110 5111 5112 5113 |
{ unsigned long pages_min = min_free_kbytes >> (PAGE_SHIFT - 10); unsigned long lowmem_pages = 0; struct zone *zone; unsigned long flags; /* Calculate total number of !ZONE_HIGHMEM pages */ for_each_zone(zone) { if (!is_highmem(zone)) |
b40da0494 mm: use zone->pre... |
5114 |
lowmem_pages += zone->managed_pages; |
1da177e4c Linux-2.6.12-rc2 |
5115 5116 5117 |
} for_each_zone(zone) { |
ac924c603 [PATCH] setup_per... |
5118 |
u64 tmp; |
1125b4e39 setup_per_zone_pa... |
5119 |
spin_lock_irqsave(&zone->lock, flags); |
b40da0494 mm: use zone->pre... |
5120 |
tmp = (u64)pages_min * zone->managed_pages; |
ac924c603 [PATCH] setup_per... |
5121 |
do_div(tmp, lowmem_pages); |
1da177e4c Linux-2.6.12-rc2 |
5122 5123 |
if (is_highmem(zone)) { /* |
669ed1752 [PATCH] mm: highm... |
5124 5125 5126 5127 |
* __GFP_HIGH and PF_MEMALLOC allocations usually don't * need highmem pages, so cap pages_min to a small * value here. * |
418589663 page allocator: u... |
5128 |
* The WMARK_HIGH-WMARK_LOW and (WMARK_LOW-WMARK_MIN) |
669ed1752 [PATCH] mm: highm... |
5129 5130 |
* deltas controls asynch page reclaim, and so should * not be capped for highmem. |
1da177e4c Linux-2.6.12-rc2 |
5131 |
*/ |
90ae8d670 mm/page_alloc.c:_... |
5132 |
unsigned long min_pages; |
1da177e4c Linux-2.6.12-rc2 |
5133 |
|
b40da0494 mm: use zone->pre... |
5134 |
min_pages = zone->managed_pages / 1024; |
90ae8d670 mm/page_alloc.c:_... |
5135 |
min_pages = clamp(min_pages, SWAP_CLUSTER_MAX, 128UL); |
418589663 page allocator: u... |
5136 |
zone->watermark[WMARK_MIN] = min_pages; |
1da177e4c Linux-2.6.12-rc2 |
5137 |
} else { |
669ed1752 [PATCH] mm: highm... |
5138 5139 |
/* * If it's a lowmem zone, reserve a number of pages |
1da177e4c Linux-2.6.12-rc2 |
5140 5141 |
* proportionate to the zone's size. */ |
418589663 page allocator: u... |
5142 |
zone->watermark[WMARK_MIN] = tmp; |
1da177e4c Linux-2.6.12-rc2 |
5143 |
} |
418589663 page allocator: u... |
5144 5145 |
zone->watermark[WMARK_LOW] = min_wmark_pages(zone) + (tmp >> 2); zone->watermark[WMARK_HIGH] = min_wmark_pages(zone) + (tmp >> 1); |
49f223a9c mm: trigger page ... |
5146 |
|
56fd56b86 Bias the location... |
5147 |
setup_zone_migrate_reserve(zone); |
1125b4e39 setup_per_zone_pa... |
5148 |
spin_unlock_irqrestore(&zone->lock, flags); |
1da177e4c Linux-2.6.12-rc2 |
5149 |
} |
cb45b0e96 [PATCH] overcommi... |
5150 5151 5152 |
/* update totalreserve_pages */ calculate_totalreserve_pages(); |
1da177e4c Linux-2.6.12-rc2 |
5153 |
} |
cfd3da1e4 mm: Serialize acc... |
5154 5155 5156 5157 5158 5159 5160 5161 5162 5163 5164 5165 5166 |
/** * setup_per_zone_wmarks - called when min_free_kbytes changes * or when memory is hot-{added|removed} * * Ensures that the watermark[min,low,high] values for each zone are set * correctly with respect to min_free_kbytes. */ void setup_per_zone_wmarks(void) { mutex_lock(&zonelists_mutex); __setup_per_zone_wmarks(); mutex_unlock(&zonelists_mutex); } |
55a4462af page_alloc: fix k... |
5167 |
/* |
556adecba vmscan: second ch... |
5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178 5179 5180 5181 5182 5183 5184 5185 5186 5187 |
* The inactive anon list should be small enough that the VM never has to * do too much work, but large enough that each inactive page has a chance * to be referenced again before it is swapped out. * * The inactive_anon ratio is the target ratio of ACTIVE_ANON to * INACTIVE_ANON pages on this zone's LRU, maintained by the * pageout code. A zone->inactive_ratio of 3 means 3:1 or 25% of * the anonymous pages are kept on the inactive list. * * total target max * memory ratio inactive anon * ------------------------------------- * 10MB 1 5MB * 100MB 1 50MB * 1GB 3 250MB * 10GB 10 0.9GB * 100GB 31 3GB * 1TB 101 10GB * 10TB 320 32GB */ |
1b79acc91 mm, mem-hotplug: ... |
5188 |
static void __meminit calculate_zone_inactive_ratio(struct zone *zone) |
556adecba vmscan: second ch... |
5189 |
{ |
96cb4df5d page-allocator: a... |
5190 |
unsigned int gb, ratio; |
556adecba vmscan: second ch... |
5191 |
|
96cb4df5d page-allocator: a... |
5192 |
/* Zone size in gigabytes */ |
b40da0494 mm: use zone->pre... |
5193 |
gb = zone->managed_pages >> (30 - PAGE_SHIFT); |
96cb4df5d page-allocator: a... |
5194 |
if (gb) |
556adecba vmscan: second ch... |
5195 |
ratio = int_sqrt(10 * gb); |
96cb4df5d page-allocator: a... |
5196 5197 |
else ratio = 1; |
556adecba vmscan: second ch... |
5198 |
|
96cb4df5d page-allocator: a... |
5199 5200 |
zone->inactive_ratio = ratio; } |
556adecba vmscan: second ch... |
5201 |
|
839a4fcc8 mm, mem-hotplug: ... |
5202 |
static void __meminit setup_per_zone_inactive_ratio(void) |
96cb4df5d page-allocator: a... |
5203 5204 5205 5206 5207 |
{ struct zone *zone; for_each_zone(zone) calculate_zone_inactive_ratio(zone); |
556adecba vmscan: second ch... |
5208 |
} |
1da177e4c Linux-2.6.12-rc2 |
5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 |
/* * Initialise min_free_kbytes. * * For small machines we want it small (128k min). For large machines * we want it large (64MB max). But it is not linear, because network * bandwidth does not increase linearly with machine size. We use * * min_free_kbytes = 4 * sqrt(lowmem_kbytes), for better accuracy: * min_free_kbytes = sqrt(lowmem_kbytes * 16) * * which yields * * 16MB: 512k * 32MB: 724k * 64MB: 1024k * 128MB: 1448k * 256MB: 2048k * 512MB: 2896k * 1024MB: 4096k * 2048MB: 5792k * 4096MB: 8192k * 8192MB: 11584k * 16384MB: 16384k */ |
1b79acc91 mm, mem-hotplug: ... |
5233 |
int __meminit init_per_zone_wmark_min(void) |
1da177e4c Linux-2.6.12-rc2 |
5234 5235 5236 5237 5238 5239 5240 5241 5242 5243 |
{ unsigned long lowmem_kbytes; lowmem_kbytes = nr_free_buffer_pages() * (PAGE_SIZE >> 10); min_free_kbytes = int_sqrt(lowmem_kbytes * 16); if (min_free_kbytes < 128) min_free_kbytes = 128; if (min_free_kbytes > 65536) min_free_kbytes = 65536; |
bc75d33f0 page-allocator: c... |
5244 |
setup_per_zone_wmarks(); |
a6cccdc36 mm, mem-hotplug: ... |
5245 |
refresh_zone_stat_thresholds(); |
1da177e4c Linux-2.6.12-rc2 |
5246 |
setup_per_zone_lowmem_reserve(); |
556adecba vmscan: second ch... |
5247 |
setup_per_zone_inactive_ratio(); |
1da177e4c Linux-2.6.12-rc2 |
5248 5249 |
return 0; } |
bc75d33f0 page-allocator: c... |
5250 |
module_init(init_per_zone_wmark_min) |
1da177e4c Linux-2.6.12-rc2 |
5251 5252 5253 5254 5255 5256 5257 |
/* * min_free_kbytes_sysctl_handler - just a wrapper around proc_dointvec() so * that we can call two helper functions whenever min_free_kbytes * changes. */ int min_free_kbytes_sysctl_handler(ctl_table *table, int write, |
8d65af789 sysctl: remove "s... |
5258 |
void __user *buffer, size_t *length, loff_t *ppos) |
1da177e4c Linux-2.6.12-rc2 |
5259 |
{ |
8d65af789 sysctl: remove "s... |
5260 |
proc_dointvec(table, write, buffer, length, ppos); |
3b1d92c56 Do not disable in... |
5261 |
if (write) |
bc75d33f0 page-allocator: c... |
5262 |
setup_per_zone_wmarks(); |
1da177e4c Linux-2.6.12-rc2 |
5263 5264 |
return 0; } |
9614634fe [PATCH] ZVC/zone_... |
5265 5266 |
#ifdef CONFIG_NUMA int sysctl_min_unmapped_ratio_sysctl_handler(ctl_table *table, int write, |
8d65af789 sysctl: remove "s... |
5267 |
void __user *buffer, size_t *length, loff_t *ppos) |
9614634fe [PATCH] ZVC/zone_... |
5268 5269 5270 |
{ struct zone *zone; int rc; |
8d65af789 sysctl: remove "s... |
5271 |
rc = proc_dointvec_minmax(table, write, buffer, length, ppos); |
9614634fe [PATCH] ZVC/zone_... |
5272 5273 5274 5275 |
if (rc) return rc; for_each_zone(zone) |
b40da0494 mm: use zone->pre... |
5276 |
zone->min_unmapped_pages = (zone->managed_pages * |
9614634fe [PATCH] ZVC/zone_... |
5277 5278 5279 |
sysctl_min_unmapped_ratio) / 100; return 0; } |
0ff38490c [PATCH] zone_recl... |
5280 5281 |
int sysctl_min_slab_ratio_sysctl_handler(ctl_table *table, int write, |
8d65af789 sysctl: remove "s... |
5282 |
void __user *buffer, size_t *length, loff_t *ppos) |
0ff38490c [PATCH] zone_recl... |
5283 5284 5285 |
{ struct zone *zone; int rc; |
8d65af789 sysctl: remove "s... |
5286 |
rc = proc_dointvec_minmax(table, write, buffer, length, ppos); |
0ff38490c [PATCH] zone_recl... |
5287 5288 5289 5290 |
if (rc) return rc; for_each_zone(zone) |
b40da0494 mm: use zone->pre... |
5291 |
zone->min_slab_pages = (zone->managed_pages * |
0ff38490c [PATCH] zone_recl... |
5292 5293 5294 |
sysctl_min_slab_ratio) / 100; return 0; } |
9614634fe [PATCH] ZVC/zone_... |
5295 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
5296 5297 5298 5299 5300 5301 |
/* * lowmem_reserve_ratio_sysctl_handler - just a wrapper around * proc_dointvec() so that we can call setup_per_zone_lowmem_reserve() * whenever sysctl_lowmem_reserve_ratio changes. * * The reserve ratio obviously has absolutely no relation with the |
418589663 page allocator: u... |
5302 |
* minimum watermarks. The lowmem reserve ratio can only make sense |
1da177e4c Linux-2.6.12-rc2 |
5303 5304 5305 |
* if in function of the boot time zone sizes. */ int lowmem_reserve_ratio_sysctl_handler(ctl_table *table, int write, |
8d65af789 sysctl: remove "s... |
5306 |
void __user *buffer, size_t *length, loff_t *ppos) |
1da177e4c Linux-2.6.12-rc2 |
5307 |
{ |
8d65af789 sysctl: remove "s... |
5308 |
proc_dointvec_minmax(table, write, buffer, length, ppos); |
1da177e4c Linux-2.6.12-rc2 |
5309 5310 5311 |
setup_per_zone_lowmem_reserve(); return 0; } |
8ad4b1fb8 [PATCH] Make high... |
5312 5313 5314 5315 5316 5317 5318 |
/* * percpu_pagelist_fraction - changes the pcp->high for each zone on each * cpu. It is the fraction of total pages in each zone that a hot per cpu pagelist * can have before it gets flushed back to buddy allocator. */ int percpu_pagelist_fraction_sysctl_handler(ctl_table *table, int write, |
8d65af789 sysctl: remove "s... |
5319 |
void __user *buffer, size_t *length, loff_t *ppos) |
8ad4b1fb8 [PATCH] Make high... |
5320 5321 5322 5323 |
{ struct zone *zone; unsigned int cpu; int ret; |
8d65af789 sysctl: remove "s... |
5324 |
ret = proc_dointvec_minmax(table, write, buffer, length, ppos); |
93278814d mm: fix division ... |
5325 |
if (!write || (ret < 0)) |
8ad4b1fb8 [PATCH] Make high... |
5326 |
return ret; |
364df0ebf mm: fix handling ... |
5327 |
for_each_populated_zone(zone) { |
99dcc3e5a this_cpu: Page al... |
5328 |
for_each_possible_cpu(cpu) { |
8ad4b1fb8 [PATCH] Make high... |
5329 |
unsigned long high; |
b40da0494 mm: use zone->pre... |
5330 |
high = zone->managed_pages / percpu_pagelist_fraction; |
99dcc3e5a this_cpu: Page al... |
5331 5332 |
setup_pagelist_highmark( per_cpu_ptr(zone->pageset, cpu), high); |
8ad4b1fb8 [PATCH] Make high... |
5333 5334 5335 5336 |
} } return 0; } |
f034b5d4e [XFRM]: Dynamic x... |
5337 |
int hashdist = HASHDIST_DEFAULT; |
1da177e4c Linux-2.6.12-rc2 |
5338 5339 5340 5341 5342 5343 5344 5345 5346 5347 5348 5349 5350 5351 5352 5353 5354 5355 5356 5357 5358 5359 5360 5361 5362 |
#ifdef CONFIG_NUMA static int __init set_hashdist(char *str) { if (!str) return 0; hashdist = simple_strtoul(str, &str, 0); return 1; } __setup("hashdist=", set_hashdist); #endif /* * allocate a large system hash table from bootmem * - it is assumed that the hash table must contain an exact power-of-2 * quantity of entries * - limit is the number of hash buckets, not the total allocation size */ void *__init alloc_large_system_hash(const char *tablename, unsigned long bucketsize, unsigned long numentries, int scale, int flags, unsigned int *_hash_shift, unsigned int *_hash_mask, |
31fe62b95 mm: add a low lim... |
5363 5364 |
unsigned long low_limit, unsigned long high_limit) |
1da177e4c Linux-2.6.12-rc2 |
5365 |
{ |
31fe62b95 mm: add a low lim... |
5366 |
unsigned long long max = high_limit; |
1da177e4c Linux-2.6.12-rc2 |
5367 5368 5369 5370 5371 5372 |
unsigned long log2qty, size; void *table = NULL; /* allow the kernel cmdline to have a say */ if (!numentries) { /* round applicable memory size up to nearest megabyte */ |
049036643 [PATCH] remove HA... |
5373 |
numentries = nr_kernel_pages; |
1da177e4c Linux-2.6.12-rc2 |
5374 5375 5376 5377 5378 5379 5380 5381 5382 |
numentries += (1UL << (20 - PAGE_SHIFT)) - 1; numentries >>= 20 - PAGE_SHIFT; numentries <<= 20 - PAGE_SHIFT; /* limit to 1 bucket per 2^scale bytes of low memory */ if (scale > PAGE_SHIFT) numentries >>= (scale - PAGE_SHIFT); else numentries <<= (PAGE_SHIFT - scale); |
9ab37b8f2 [PATCH] Sanely si... |
5383 5384 |
/* Make sure we've got at least a 0-order allocation.. */ |
2c85f51d2 mm: also use allo... |
5385 5386 5387 5388 5389 5390 5391 5392 |
if (unlikely(flags & HASH_SMALL)) { /* Makes no sense without HASH_EARLY */ WARN_ON(!(flags & HASH_EARLY)); if (!(numentries >> *_hash_shift)) { numentries = 1UL << *_hash_shift; BUG_ON(!numentries); } } else if (unlikely((numentries * bucketsize) < PAGE_SIZE)) |
9ab37b8f2 [PATCH] Sanely si... |
5393 |
numentries = PAGE_SIZE / bucketsize; |
1da177e4c Linux-2.6.12-rc2 |
5394 |
} |
6e692ed37 [PATCH] fix alloc... |
5395 |
numentries = roundup_pow_of_two(numentries); |
1da177e4c Linux-2.6.12-rc2 |
5396 5397 5398 5399 5400 5401 |
/* limit allocation size to 1/16 total memory by default */ if (max == 0) { max = ((unsigned long long)nr_all_pages << PAGE_SHIFT) >> 4; do_div(max, bucketsize); } |
074b85175 vfs: fix panic in... |
5402 |
max = min(max, 0x80000000ULL); |
1da177e4c Linux-2.6.12-rc2 |
5403 |
|
31fe62b95 mm: add a low lim... |
5404 5405 |
if (numentries < low_limit) numentries = low_limit; |
1da177e4c Linux-2.6.12-rc2 |
5406 5407 |
if (numentries > max) numentries = max; |
f0d1b0b30 [PATCH] LOG2: Imp... |
5408 |
log2qty = ilog2(numentries); |
1da177e4c Linux-2.6.12-rc2 |
5409 5410 5411 5412 |
do { size = bucketsize << log2qty; if (flags & HASH_EARLY) |
74768ed83 page allocator: u... |
5413 |
table = alloc_bootmem_nopanic(size); |
1da177e4c Linux-2.6.12-rc2 |
5414 5415 5416 |
else if (hashdist) table = __vmalloc(size, GFP_ATOMIC, PAGE_KERNEL); else { |
1037b83bd MM: alloc_large_s... |
5417 5418 |
/* * If bucketsize is not a power-of-two, we may free |
a1dd268cf mm: use alloc_pag... |
5419 5420 |
* some pages at the end of hash table which * alloc_pages_exact() automatically does |
1037b83bd MM: alloc_large_s... |
5421 |
*/ |
264ef8a90 kmemleak: Remove ... |
5422 |
if (get_order(size) < MAX_ORDER) { |
a1dd268cf mm: use alloc_pag... |
5423 |
table = alloc_pages_exact(size, GFP_ATOMIC); |
264ef8a90 kmemleak: Remove ... |
5424 5425 |
kmemleak_alloc(table, size, 1, GFP_ATOMIC); } |
1da177e4c Linux-2.6.12-rc2 |
5426 5427 5428 5429 5430 5431 |
} } while (!table && size > PAGE_SIZE && --log2qty); if (!table) panic("Failed to allocate %s hash table ", tablename); |
f241e6607 mm: alloc_large_s... |
5432 5433 |
printk(KERN_INFO "%s hash table entries: %ld (order: %d, %lu bytes) ", |
1da177e4c Linux-2.6.12-rc2 |
5434 |
tablename, |
f241e6607 mm: alloc_large_s... |
5435 |
(1UL << log2qty), |
f0d1b0b30 [PATCH] LOG2: Imp... |
5436 |
ilog2(size) - PAGE_SHIFT, |
1da177e4c Linux-2.6.12-rc2 |
5437 5438 5439 5440 5441 5442 5443 5444 5445 |
size); if (_hash_shift) *_hash_shift = log2qty; if (_hash_mask) *_hash_mask = (1 << log2qty) - 1; return table; } |
a117e66ed [PATCH] unify pfn... |
5446 |
|
835c134ec Add a bitmap that... |
5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457 5458 5459 5460 5461 |
/* Return a pointer to the bitmap storing bits affecting a block of pages */ static inline unsigned long *get_pageblock_bitmap(struct zone *zone, unsigned long pfn) { #ifdef CONFIG_SPARSEMEM return __pfn_to_section(pfn)->pageblock_flags; #else return zone->pageblock_flags; #endif /* CONFIG_SPARSEMEM */ } static inline int pfn_to_bitidx(struct zone *zone, unsigned long pfn) { #ifdef CONFIG_SPARSEMEM pfn &= (PAGES_PER_SECTION-1); |
d9c234005 Do not depend on ... |
5462 |
return (pfn >> pageblock_order) * NR_PAGEBLOCK_BITS; |
835c134ec Add a bitmap that... |
5463 |
#else |
c060f943d mm: use aligned z... |
5464 |
pfn = pfn - round_down(zone->zone_start_pfn, pageblock_nr_pages); |
d9c234005 Do not depend on ... |
5465 |
return (pfn >> pageblock_order) * NR_PAGEBLOCK_BITS; |
835c134ec Add a bitmap that... |
5466 5467 5468 5469 |
#endif /* CONFIG_SPARSEMEM */ } /** |
d9c234005 Do not depend on ... |
5470 |
* get_pageblock_flags_group - Return the requested group of flags for the pageblock_nr_pages block of pages |
835c134ec Add a bitmap that... |
5471 5472 5473 5474 5475 5476 5477 5478 5479 5480 5481 5482 5483 5484 5485 5486 5487 5488 5489 5490 5491 5492 |
* @page: The page within the block of interest * @start_bitidx: The first bit of interest to retrieve * @end_bitidx: The last bit of interest * returns pageblock_bits flags */ unsigned long get_pageblock_flags_group(struct page *page, int start_bitidx, int end_bitidx) { struct zone *zone; unsigned long *bitmap; unsigned long pfn, bitidx; unsigned long flags = 0; unsigned long value = 1; zone = page_zone(page); pfn = page_to_pfn(page); bitmap = get_pageblock_bitmap(zone, pfn); bitidx = pfn_to_bitidx(zone, pfn); for (; start_bitidx <= end_bitidx; start_bitidx++, value <<= 1) if (test_bit(bitidx + start_bitidx, bitmap)) flags |= value; |
6220ec784 [PATCH] highest_p... |
5493 |
|
835c134ec Add a bitmap that... |
5494 5495 5496 5497 |
return flags; } /** |
d9c234005 Do not depend on ... |
5498 |
* set_pageblock_flags_group - Set the requested group of flags for a pageblock_nr_pages block of pages |
835c134ec Add a bitmap that... |
5499 5500 5501 5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512 5513 5514 5515 |
* @page: The page within the block of interest * @start_bitidx: The first bit of interest * @end_bitidx: The last bit of interest * @flags: The flags to set */ void set_pageblock_flags_group(struct page *page, unsigned long flags, int start_bitidx, int end_bitidx) { struct zone *zone; unsigned long *bitmap; unsigned long pfn, bitidx; unsigned long value = 1; zone = page_zone(page); pfn = page_to_pfn(page); bitmap = get_pageblock_bitmap(zone, pfn); bitidx = pfn_to_bitidx(zone, pfn); |
108bcc96e mm: add & use zon... |
5516 |
VM_BUG_ON(!zone_spans_pfn(zone, pfn)); |
835c134ec Add a bitmap that... |
5517 5518 5519 5520 5521 5522 5523 |
for (; start_bitidx <= end_bitidx; start_bitidx++, value <<= 1) if (flags & value) __set_bit(bitidx + start_bitidx, bitmap); else __clear_bit(bitidx + start_bitidx, bitmap); } |
a5d76b54a memory unplug: pa... |
5524 5525 |
/* |
80934513b mm: clean up __co... |
5526 5527 5528 5529 5530 5531 |
* This function checks whether pageblock includes unmovable pages or not. * If @count is not zero, it is okay to include less @count unmovable pages * * PageLRU check wihtout isolation or lru_lock could race so that * MIGRATE_MOVABLE block might include unmovable pages. It means you can't * expect this function should be exact. |
a5d76b54a memory unplug: pa... |
5532 |
*/ |
b023f4681 memory-hotplug: s... |
5533 5534 |
bool has_unmovable_pages(struct zone *zone, struct page *page, int count, bool skip_hwpoisoned_pages) |
49ac82558 memory hotplug: u... |
5535 5536 |
{ unsigned long pfn, iter, found; |
47118af07 mm: mmzone: MIGRA... |
5537 |
int mt; |
49ac82558 memory hotplug: u... |
5538 5539 |
/* * For avoiding noise data, lru_add_drain_all() should be called |
80934513b mm: clean up __co... |
5540 |
* If ZONE_MOVABLE, the zone never contains unmovable pages |
49ac82558 memory hotplug: u... |
5541 5542 |
*/ if (zone_idx(zone) == ZONE_MOVABLE) |
80934513b mm: clean up __co... |
5543 |
return false; |
47118af07 mm: mmzone: MIGRA... |
5544 5545 |
mt = get_pageblock_migratetype(page); if (mt == MIGRATE_MOVABLE || is_migrate_cma(mt)) |
80934513b mm: clean up __co... |
5546 |
return false; |
49ac82558 memory hotplug: u... |
5547 5548 5549 5550 |
pfn = page_to_pfn(page); for (found = 0, iter = 0; iter < pageblock_nr_pages; iter++) { unsigned long check = pfn + iter; |
29723fccc mm: fix dubious c... |
5551 |
if (!pfn_valid_within(check)) |
49ac82558 memory hotplug: u... |
5552 |
continue; |
29723fccc mm: fix dubious c... |
5553 |
|
49ac82558 memory hotplug: u... |
5554 |
page = pfn_to_page(check); |
97d255c81 mm: do not use pa... |
5555 5556 5557 5558 5559 5560 5561 |
/* * We can't use page_count without pin a page * because another CPU can free compound page. * This check already skips compound tails of THP * because their page->_count is zero at all time. */ if (!atomic_read(&page->_count)) { |
49ac82558 memory hotplug: u... |
5562 5563 5564 5565 |
if (PageBuddy(page)) iter += (1 << page_order(page)) - 1; continue; } |
97d255c81 mm: do not use pa... |
5566 |
|
b023f4681 memory-hotplug: s... |
5567 5568 5569 5570 5571 5572 |
/* * The HWPoisoned page may be not in buddy system, and * page_count() is not 0. */ if (skip_hwpoisoned_pages && PageHWPoison(page)) continue; |
49ac82558 memory hotplug: u... |
5573 5574 5575 5576 5577 5578 5579 5580 5581 5582 5583 5584 5585 5586 5587 5588 |
if (!PageLRU(page)) found++; /* * If there are RECLAIMABLE pages, we need to check it. * But now, memory offline itself doesn't call shrink_slab() * and it still to be fixed. */ /* * If the page is not RAM, page_count()should be 0. * we don't need more check. This is an _used_ not-movable page. * * The problematic thing here is PG_reserved pages. PG_reserved * is set to both of a memory hole page and a _used_ kernel * page at boot. */ if (found > count) |
80934513b mm: clean up __co... |
5589 |
return true; |
49ac82558 memory hotplug: u... |
5590 |
} |
80934513b mm: clean up __co... |
5591 |
return false; |
49ac82558 memory hotplug: u... |
5592 5593 5594 5595 |
} bool is_pageblock_removable_nolock(struct page *page) { |
656a07062 mm: __count_immob... |
5596 5597 |
struct zone *zone; unsigned long pfn; |
687875fb7 mm: fix NULL ptr ... |
5598 5599 5600 5601 5602 |
/* * We have to be careful here because we are iterating over memory * sections which are not zone aware so we might end up outside of * the zone but still within the section. |
656a07062 mm: __count_immob... |
5603 5604 |
* We have to take care about the node as well. If the node is offline * its NODE_DATA will be NULL - see page_zone. |
687875fb7 mm: fix NULL ptr ... |
5605 |
*/ |
656a07062 mm: __count_immob... |
5606 5607 5608 5609 5610 |
if (!node_online(page_to_nid(page))) return false; zone = page_zone(page); pfn = page_to_pfn(page); |
108bcc96e mm: add & use zon... |
5611 |
if (!zone_spans_pfn(zone, pfn)) |
687875fb7 mm: fix NULL ptr ... |
5612 |
return false; |
b023f4681 memory-hotplug: s... |
5613 |
return !has_unmovable_pages(zone, page, 0, true); |
a5d76b54a memory unplug: pa... |
5614 |
} |
0c0e61958 memory unplug: pa... |
5615 |
|
041d3a8cd mm: page_alloc: i... |
5616 5617 5618 5619 5620 5621 5622 5623 5624 5625 5626 5627 5628 |
#ifdef CONFIG_CMA static unsigned long pfn_max_align_down(unsigned long pfn) { return pfn & ~(max_t(unsigned long, MAX_ORDER_NR_PAGES, pageblock_nr_pages) - 1); } static unsigned long pfn_max_align_up(unsigned long pfn) { return ALIGN(pfn, max_t(unsigned long, MAX_ORDER_NR_PAGES, pageblock_nr_pages)); } |
041d3a8cd mm: page_alloc: i... |
5629 |
/* [start, end) must belong to a single zone. */ |
bb13ffeb9 mm: compaction: c... |
5630 5631 |
static int __alloc_contig_migrate_range(struct compact_control *cc, unsigned long start, unsigned long end) |
041d3a8cd mm: page_alloc: i... |
5632 5633 |
{ /* This function is based on compact_zone() from compaction.c. */ |
beb51eaa8 cma: decrease cc.... |
5634 |
unsigned long nr_reclaimed; |
041d3a8cd mm: page_alloc: i... |
5635 5636 5637 |
unsigned long pfn = start; unsigned int tries = 0; int ret = 0; |
be49a6e13 mm: use migrate_p... |
5638 |
migrate_prep(); |
041d3a8cd mm: page_alloc: i... |
5639 |
|
bb13ffeb9 mm: compaction: c... |
5640 |
while (pfn < end || !list_empty(&cc->migratepages)) { |
041d3a8cd mm: page_alloc: i... |
5641 5642 5643 5644 |
if (fatal_signal_pending(current)) { ret = -EINTR; break; } |
bb13ffeb9 mm: compaction: c... |
5645 5646 5647 |
if (list_empty(&cc->migratepages)) { cc->nr_migratepages = 0; pfn = isolate_migratepages_range(cc->zone, cc, |
e46a28790 CMA: migrate mloc... |
5648 |
pfn, end, true); |
041d3a8cd mm: page_alloc: i... |
5649 5650 5651 5652 5653 5654 5655 5656 5657 |
if (!pfn) { ret = -EINTR; break; } tries = 0; } else if (++tries == 5) { ret = ret < 0 ? ret : -EBUSY; break; } |
beb51eaa8 cma: decrease cc.... |
5658 5659 5660 |
nr_reclaimed = reclaim_clean_pages_from_list(cc->zone, &cc->migratepages); cc->nr_migratepages -= nr_reclaimed; |
02c6de8d7 mm: cma: discard ... |
5661 |
|
9c620e2bc mm: remove offlin... |
5662 5663 |
ret = migrate_pages(&cc->migratepages, alloc_migrate_target, 0, MIGRATE_SYNC, MR_CMA); |
041d3a8cd mm: page_alloc: i... |
5664 |
} |
2a6f51241 CMA: make putback... |
5665 5666 5667 5668 5669 |
if (ret < 0) { putback_movable_pages(&cc->migratepages); return ret; } return 0; |
041d3a8cd mm: page_alloc: i... |
5670 5671 5672 5673 5674 5675 |
} /** * alloc_contig_range() -- tries to allocate given range of pages * @start: start PFN to allocate * @end: one-past-the-last PFN to allocate |
0815f3d81 mm: page_isolatio... |
5676 5677 5678 5679 |
* @migratetype: migratetype of the underlaying pageblocks (either * #MIGRATE_MOVABLE or #MIGRATE_CMA). All pageblocks * in range must have the same migratetype and it must * be either of the two. |
041d3a8cd mm: page_alloc: i... |
5680 5681 5682 5683 5684 5685 5686 5687 5688 5689 5690 5691 |
* * The PFN range does not have to be pageblock or MAX_ORDER_NR_PAGES * aligned, however it's the caller's responsibility to guarantee that * we are the only thread that changes migrate type of pageblocks the * pages fall in. * * The PFN range must belong to a single zone. * * Returns zero on success or negative error code. On success all * pages which PFN is in [start, end) are allocated for the caller and * need to be freed with free_contig_range(). */ |
0815f3d81 mm: page_isolatio... |
5692 5693 |
int alloc_contig_range(unsigned long start, unsigned long end, unsigned migratetype) |
041d3a8cd mm: page_alloc: i... |
5694 |
{ |
041d3a8cd mm: page_alloc: i... |
5695 5696 |
unsigned long outer_start, outer_end; int ret = 0, order; |
bb13ffeb9 mm: compaction: c... |
5697 5698 5699 5700 5701 5702 5703 5704 |
struct compact_control cc = { .nr_migratepages = 0, .order = -1, .zone = page_zone(pfn_to_page(start)), .sync = true, .ignore_skip_hint = true, }; INIT_LIST_HEAD(&cc.migratepages); |
041d3a8cd mm: page_alloc: i... |
5705 5706 5707 5708 5709 5710 5711 5712 5713 5714 5715 5716 5717 5718 5719 5720 5721 5722 5723 5724 5725 5726 5727 5728 5729 |
/* * What we do here is we mark all pageblocks in range as * MIGRATE_ISOLATE. Because pageblock and max order pages may * have different sizes, and due to the way page allocator * work, we align the range to biggest of the two pages so * that page allocator won't try to merge buddies from * different pageblocks and change MIGRATE_ISOLATE to some * other migration type. * * Once the pageblocks are marked as MIGRATE_ISOLATE, we * migrate the pages from an unaligned range (ie. pages that * we are interested in). This will put all the pages in * range back to page allocator as MIGRATE_ISOLATE. * * When this is done, we take the pages in range from page * allocator removing them from the buddy system. This way * page allocator will never consider using them. * * This lets us mark the pageblocks back as * MIGRATE_CMA/MIGRATE_MOVABLE so that free pages in the * aligned range but not in the unaligned, original range are * put back to page allocator so that buddy can use them. */ ret = start_isolate_page_range(pfn_max_align_down(start), |
b023f4681 memory-hotplug: s... |
5730 5731 |
pfn_max_align_up(end), migratetype, false); |
041d3a8cd mm: page_alloc: i... |
5732 |
if (ret) |
86a595f96 mm/page_alloc.c:a... |
5733 |
return ret; |
041d3a8cd mm: page_alloc: i... |
5734 |
|
bb13ffeb9 mm: compaction: c... |
5735 |
ret = __alloc_contig_migrate_range(&cc, start, end); |
041d3a8cd mm: page_alloc: i... |
5736 5737 5738 5739 5740 5741 5742 5743 5744 5745 5746 5747 5748 5749 5750 5751 5752 5753 5754 5755 5756 5757 5758 5759 5760 5761 5762 5763 5764 5765 5766 5767 5768 5769 |
if (ret) goto done; /* * Pages from [start, end) are within a MAX_ORDER_NR_PAGES * aligned blocks that are marked as MIGRATE_ISOLATE. What's * more, all pages in [start, end) are free in page allocator. * What we are going to do is to allocate all pages from * [start, end) (that is remove them from page allocator). * * The only problem is that pages at the beginning and at the * end of interesting range may be not aligned with pages that * page allocator holds, ie. they can be part of higher order * pages. Because of this, we reserve the bigger range and * once this is done free the pages we are not interested in. * * We don't have to hold zone->lock here because the pages are * isolated thus they won't get removed from buddy. */ lru_add_drain_all(); drain_all_pages(); order = 0; outer_start = start; while (!PageBuddy(pfn_to_page(outer_start))) { if (++order >= MAX_ORDER) { ret = -EBUSY; goto done; } outer_start &= ~0UL << order; } /* Make sure the range is really isolated. */ |
b023f4681 memory-hotplug: s... |
5770 |
if (test_pages_isolated(outer_start, end, false)) { |
041d3a8cd mm: page_alloc: i... |
5771 5772 5773 5774 5775 5776 |
pr_warn("alloc_contig_range test_pages_isolated(%lx, %lx) failed ", outer_start, end); ret = -EBUSY; goto done; } |
49f223a9c mm: trigger page ... |
5777 5778 |
/* Grab isolated pages from freelists. */ |
bb13ffeb9 mm: compaction: c... |
5779 |
outer_end = isolate_freepages_range(&cc, outer_start, end); |
041d3a8cd mm: page_alloc: i... |
5780 5781 5782 5783 5784 5785 5786 5787 5788 5789 5790 5791 5792 |
if (!outer_end) { ret = -EBUSY; goto done; } /* Free head and tail (if any) */ if (start != outer_start) free_contig_range(outer_start, start - outer_start); if (end != outer_end) free_contig_range(end, outer_end - end); done: undo_isolate_page_range(pfn_max_align_down(start), |
0815f3d81 mm: page_isolatio... |
5793 |
pfn_max_align_up(end), migratetype); |
041d3a8cd mm: page_alloc: i... |
5794 5795 5796 5797 5798 |
return ret; } void free_contig_range(unsigned long pfn, unsigned nr_pages) { |
bcc2b02f4 mm: cma: WARN if ... |
5799 5800 5801 5802 5803 5804 5805 5806 5807 5808 |
unsigned int count = 0; for (; nr_pages--; pfn++) { struct page *page = pfn_to_page(pfn); count += page_count(page) != 1; __free_page(page); } WARN(count != 0, "%d pages are still in use! ", count); |
041d3a8cd mm: page_alloc: i... |
5809 5810 |
} #endif |
4ed7e0222 mm/hotplug: mark ... |
5811 5812 5813 5814 5815 5816 5817 5818 5819 5820 5821 5822 5823 5824 5825 5826 5827 |
#ifdef CONFIG_MEMORY_HOTPLUG static int __meminit __zone_pcp_update(void *data) { struct zone *zone = data; int cpu; unsigned long batch = zone_batchsize(zone), flags; for_each_possible_cpu(cpu) { struct per_cpu_pageset *pset; struct per_cpu_pages *pcp; pset = per_cpu_ptr(zone->pageset, cpu); pcp = &pset->pcp; local_irq_save(flags); if (pcp->count > 0) free_pcppages_bulk(zone, pcp->count, pcp); |
5a8838138 memory-hotplug: f... |
5828 |
drain_zonestat(zone, pset); |
4ed7e0222 mm/hotplug: mark ... |
5829 5830 5831 5832 5833 5834 5835 5836 5837 5838 5839 |
setup_pageset(pset, batch); local_irq_restore(flags); } return 0; } void __meminit zone_pcp_update(struct zone *zone) { stop_machine(__zone_pcp_update, zone, NULL); } #endif |
340175b7d mm/hotplug: free ... |
5840 5841 5842 |
void zone_pcp_reset(struct zone *zone) { unsigned long flags; |
5a8838138 memory-hotplug: f... |
5843 5844 |
int cpu; struct per_cpu_pageset *pset; |
340175b7d mm/hotplug: free ... |
5845 5846 5847 5848 |
/* avoid races with drain_pages() */ local_irq_save(flags); if (zone->pageset != &boot_pageset) { |
5a8838138 memory-hotplug: f... |
5849 5850 5851 5852 |
for_each_online_cpu(cpu) { pset = per_cpu_ptr(zone->pageset, cpu); drain_zonestat(zone, pset); } |
340175b7d mm/hotplug: free ... |
5853 5854 5855 5856 5857 |
free_percpu(zone->pageset); zone->pageset = &boot_pageset; } local_irq_restore(flags); } |
6dcd73d70 memory-hotplug: a... |
5858 |
#ifdef CONFIG_MEMORY_HOTREMOVE |
0c0e61958 memory unplug: pa... |
5859 5860 5861 5862 5863 5864 5865 5866 5867 5868 5869 5870 5871 5872 5873 5874 5875 5876 5877 5878 5879 5880 5881 5882 5883 5884 |
/* * All pages in the range must be isolated before calling this. */ void __offline_isolated_pages(unsigned long start_pfn, unsigned long end_pfn) { struct page *page; struct zone *zone; int order, i; unsigned long pfn; unsigned long flags; /* find the first valid pfn */ for (pfn = start_pfn; pfn < end_pfn; pfn++) if (pfn_valid(pfn)) break; if (pfn == end_pfn) return; zone = page_zone(pfn_to_page(pfn)); spin_lock_irqsave(&zone->lock, flags); pfn = start_pfn; while (pfn < end_pfn) { if (!pfn_valid(pfn)) { pfn++; continue; } page = pfn_to_page(pfn); |
b023f4681 memory-hotplug: s... |
5885 5886 5887 5888 5889 5890 5891 5892 5893 |
/* * The HWPoisoned page may be not in buddy system, and * page_count() is not 0. */ if (unlikely(!PageBuddy(page) && PageHWPoison(page))) { pfn++; SetPageReserved(page); continue; } |
0c0e61958 memory unplug: pa... |
5894 5895 5896 5897 5898 5899 5900 5901 5902 5903 5904 |
BUG_ON(page_count(page)); BUG_ON(!PageBuddy(page)); order = page_order(page); #ifdef CONFIG_DEBUG_VM printk(KERN_INFO "remove from free list %lx %d %lx ", pfn, 1 << order, end_pfn); #endif list_del(&page->lru); rmv_page_order(page); zone->free_area[order].nr_free--; |
764225780 mm/memory-hotplug... |
5905 5906 5907 5908 |
#ifdef CONFIG_HIGHMEM if (PageHighMem(page)) totalhigh_pages -= 1 << order; #endif |
0c0e61958 memory unplug: pa... |
5909 5910 5911 5912 5913 5914 5915 |
for (i = 0; i < (1 << order); i++) SetPageReserved((page+i)); pfn += (1 << order); } spin_unlock_irqrestore(&zone->lock, flags); } #endif |
8d22ba1b7 HWPOISON: detect ... |
5916 5917 5918 5919 5920 5921 5922 5923 5924 5925 5926 5927 5928 5929 5930 5931 5932 5933 5934 5935 5936 |
#ifdef CONFIG_MEMORY_FAILURE bool is_free_buddy_page(struct page *page) { struct zone *zone = page_zone(page); unsigned long pfn = page_to_pfn(page); unsigned long flags; int order; spin_lock_irqsave(&zone->lock, flags); for (order = 0; order < MAX_ORDER; order++) { struct page *page_head = page - (pfn & ((1 << order) - 1)); if (PageBuddy(page_head) && page_order(page_head) >= order) break; } spin_unlock_irqrestore(&zone->lock, flags); return order < MAX_ORDER; } #endif |
718a38211 mm: introduce dum... |
5937 |
|
51300cef4 mm/page_alloc.c: ... |
5938 |
static const struct trace_print_flags pageflag_names[] = { |
718a38211 mm: introduce dum... |
5939 5940 5941 5942 5943 5944 5945 5946 5947 5948 5949 5950 5951 5952 5953 5954 5955 5956 5957 5958 5959 5960 5961 |
{1UL << PG_locked, "locked" }, {1UL << PG_error, "error" }, {1UL << PG_referenced, "referenced" }, {1UL << PG_uptodate, "uptodate" }, {1UL << PG_dirty, "dirty" }, {1UL << PG_lru, "lru" }, {1UL << PG_active, "active" }, {1UL << PG_slab, "slab" }, {1UL << PG_owner_priv_1, "owner_priv_1" }, {1UL << PG_arch_1, "arch_1" }, {1UL << PG_reserved, "reserved" }, {1UL << PG_private, "private" }, {1UL << PG_private_2, "private_2" }, {1UL << PG_writeback, "writeback" }, #ifdef CONFIG_PAGEFLAGS_EXTENDED {1UL << PG_head, "head" }, {1UL << PG_tail, "tail" }, #else {1UL << PG_compound, "compound" }, #endif {1UL << PG_swapcache, "swapcache" }, {1UL << PG_mappedtodisk, "mappedtodisk" }, {1UL << PG_reclaim, "reclaim" }, |
718a38211 mm: introduce dum... |
5962 5963 5964 5965 5966 5967 5968 5969 5970 5971 5972 |
{1UL << PG_swapbacked, "swapbacked" }, {1UL << PG_unevictable, "unevictable" }, #ifdef CONFIG_MMU {1UL << PG_mlocked, "mlocked" }, #endif #ifdef CONFIG_ARCH_USES_PG_UNCACHED {1UL << PG_uncached, "uncached" }, #endif #ifdef CONFIG_MEMORY_FAILURE {1UL << PG_hwpoison, "hwpoison" }, #endif |
be9cd873e mm/buddy: dump PG... |
5973 5974 5975 |
#ifdef CONFIG_TRANSPARENT_HUGEPAGE {1UL << PG_compound_lock, "compound_lock" }, #endif |
718a38211 mm: introduce dum... |
5976 5977 5978 5979 5980 5981 5982 |
}; static void dump_page_flags(unsigned long flags) { const char *delim = ""; unsigned long mask; int i; |
51300cef4 mm/page_alloc.c: ... |
5983 |
BUILD_BUG_ON(ARRAY_SIZE(pageflag_names) != __NR_PAGEFLAGS); |
acc50c110 mm: page_alloc: c... |
5984 |
|
718a38211 mm: introduce dum... |
5985 5986 5987 5988 |
printk(KERN_ALERT "page flags: %#lx(", flags); /* remove zone id */ flags &= (1UL << NR_PAGEFLAGS) - 1; |
51300cef4 mm/page_alloc.c: ... |
5989 |
for (i = 0; i < ARRAY_SIZE(pageflag_names) && flags; i++) { |
718a38211 mm: introduce dum... |
5990 5991 5992 5993 5994 5995 5996 5997 5998 5999 6000 6001 6002 6003 6004 6005 6006 6007 6008 6009 6010 6011 6012 |
mask = pageflag_names[i].mask; if ((flags & mask) != mask) continue; flags &= ~mask; printk("%s%s", delim, pageflag_names[i].name); delim = "|"; } /* check for left over flags */ if (flags) printk("%s%#lx", delim, flags); printk(") "); } void dump_page(struct page *page) { printk(KERN_ALERT "page:%p count:%d mapcount:%d mapping:%p index:%#lx ", |
4e9f64c42 thp: fix bad_page... |
6013 |
page, atomic_read(&page->_count), page_mapcount(page), |
718a38211 mm: introduce dum... |
6014 6015 |
page->mapping, page->index); dump_page_flags(page->flags); |
f212ad7cf memcg: add memcg ... |
6016 |
mem_cgroup_print_bad_page(page); |
718a38211 mm: introduce dum... |
6017 |
} |