Blame view
mm/page_alloc.c
83.2 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 21 22 |
#include <linux/stddef.h> #include <linux/mm.h> #include <linux/swap.h> #include <linux/interrupt.h> #include <linux/pagemap.h> #include <linux/bootmem.h> #include <linux/compiler.h> |
9f1583339 [PATCH] use add_t... |
23 |
#include <linux/kernel.h> |
1da177e4c Linux-2.6.12-rc2 |
24 25 26 27 28 29 30 31 32 33 |
#include <linux/module.h> #include <linux/suspend.h> #include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/slab.h> #include <linux/notifier.h> #include <linux/topology.h> #include <linux/sysctl.h> #include <linux/cpu.h> #include <linux/cpuset.h> |
bdc8cb984 [PATCH] memory ho... |
34 |
#include <linux/memory_hotplug.h> |
1da177e4c Linux-2.6.12-rc2 |
35 36 |
#include <linux/nodemask.h> #include <linux/vmalloc.h> |
4be38e351 [PATCH] mm: move ... |
37 |
#include <linux/mempolicy.h> |
6811378e7 [PATCH] wait_tabl... |
38 |
#include <linux/stop_machine.h> |
c713216de [PATCH] Introduce... |
39 40 |
#include <linux/sort.h> #include <linux/pfn.h> |
1da177e4c Linux-2.6.12-rc2 |
41 42 |
#include <asm/tlbflush.h> |
ac924c603 [PATCH] setup_per... |
43 |
#include <asm/div64.h> |
1da177e4c Linux-2.6.12-rc2 |
44 45 46 47 48 49 |
#include "internal.h" /* * MCD - HACK: Find somewhere to initialize this EARLY, or make this * initializer cleaner */ |
c3d8c1414 [PATCH] More __re... |
50 |
nodemask_t node_online_map __read_mostly = { { [0] = 1UL } }; |
7223a93a5 [IA64] Export nod... |
51 |
EXPORT_SYMBOL(node_online_map); |
c3d8c1414 [PATCH] More __re... |
52 |
nodemask_t node_possible_map __read_mostly = NODE_MASK_ALL; |
7223a93a5 [IA64] Export nod... |
53 |
EXPORT_SYMBOL(node_possible_map); |
6c231b7ba [PATCH] Additions... |
54 |
unsigned long totalram_pages __read_mostly; |
cb45b0e96 [PATCH] overcommi... |
55 |
unsigned long totalreserve_pages __read_mostly; |
1da177e4c Linux-2.6.12-rc2 |
56 |
long nr_swap_pages; |
8ad4b1fb8 [PATCH] Make high... |
57 |
int percpu_pagelist_fraction; |
1da177e4c Linux-2.6.12-rc2 |
58 |
|
d98c7a098 [PATCH] compound ... |
59 |
static void __free_pages_ok(struct page *page, unsigned int order); |
a226f6c89 [PATCH] FRV: Clea... |
60 |
|
1da177e4c Linux-2.6.12-rc2 |
61 62 63 64 65 66 67 |
/* * 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... |
68 69 70 |
* * 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 |
71 |
*/ |
2f1b62486 [PATCH] reduce MA... |
72 73 |
int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = { 256, |
fb0e7942b [PATCH] reduce MA... |
74 |
#ifdef CONFIG_ZONE_DMA32 |
2f1b62486 [PATCH] reduce MA... |
75 |
256, |
fb0e7942b [PATCH] reduce MA... |
76 |
#endif |
e53ef38d0 [PATCH] reduce MA... |
77 |
#ifdef CONFIG_HIGHMEM |
2f1b62486 [PATCH] reduce MA... |
78 |
32 |
e53ef38d0 [PATCH] reduce MA... |
79 |
#endif |
2f1b62486 [PATCH] reduce MA... |
80 |
}; |
1da177e4c Linux-2.6.12-rc2 |
81 82 |
EXPORT_SYMBOL(totalram_pages); |
1da177e4c Linux-2.6.12-rc2 |
83 84 85 86 87 |
/* * Used by page_zone() to look up the address of the struct zone whose * id is encoded in the upper bits of page->flags */ |
c3d8c1414 [PATCH] More __re... |
88 |
struct zone *zone_table[1 << ZONETABLE_SHIFT] __read_mostly; |
1da177e4c Linux-2.6.12-rc2 |
89 |
EXPORT_SYMBOL(zone_table); |
2f1b62486 [PATCH] reduce MA... |
90 91 |
static char *zone_names[MAX_NR_ZONES] = { "DMA", |
fb0e7942b [PATCH] reduce MA... |
92 |
#ifdef CONFIG_ZONE_DMA32 |
2f1b62486 [PATCH] reduce MA... |
93 |
"DMA32", |
fb0e7942b [PATCH] reduce MA... |
94 |
#endif |
2f1b62486 [PATCH] reduce MA... |
95 |
"Normal", |
e53ef38d0 [PATCH] reduce MA... |
96 |
#ifdef CONFIG_HIGHMEM |
2f1b62486 [PATCH] reduce MA... |
97 |
"HighMem" |
e53ef38d0 [PATCH] reduce MA... |
98 |
#endif |
2f1b62486 [PATCH] reduce MA... |
99 |
}; |
1da177e4c Linux-2.6.12-rc2 |
100 |
int min_free_kbytes = 1024; |
86356ab14 [PATCH] wait_tabl... |
101 102 |
unsigned long __meminitdata nr_kernel_pages; unsigned long __meminitdata nr_all_pages; |
0e0b864e0 [PATCH] Account f... |
103 |
static unsigned long __initdata dma_reserve; |
1da177e4c Linux-2.6.12-rc2 |
104 |
|
c713216de [PATCH] Introduce... |
105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 |
#ifdef CONFIG_ARCH_POPULATES_NODE_MAP /* * MAX_ACTIVE_REGIONS determines the maxmimum number of distinct * ranges of memory (RAM) that may be registered with add_active_range(). * Ranges passed to add_active_range() will be merged if possible * so the number of times add_active_range() can be called is * related to the number of nodes and the number of holes */ #ifdef CONFIG_MAX_ACTIVE_REGIONS /* Allow an architecture to set MAX_ACTIVE_REGIONS to save memory */ #define MAX_ACTIVE_REGIONS CONFIG_MAX_ACTIVE_REGIONS #else #if MAX_NUMNODES >= 32 /* If there can be many nodes, allow up to 50 holes per node */ #define MAX_ACTIVE_REGIONS (MAX_NUMNODES*50) #else /* By default, allow up to 256 distinct regions */ #define MAX_ACTIVE_REGIONS 256 #endif #endif struct node_active_region __initdata early_node_map[MAX_ACTIVE_REGIONS]; int __initdata nr_nodemap_entries; unsigned long __initdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES]; unsigned long __initdata arch_zone_highest_possible_pfn[MAX_NR_ZONES]; |
fb01439c5 [PATCH] Allow an ... |
130 131 132 133 |
#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE unsigned long __initdata node_boundary_start_pfn[MAX_NUMNODES]; unsigned long __initdata node_boundary_end_pfn[MAX_NUMNODES]; #endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */ |
c713216de [PATCH] Introduce... |
134 |
#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */ |
13e7444b0 [PATCH] mm: remov... |
135 |
#ifdef CONFIG_DEBUG_VM |
c6a57e19e [PATCH] memory ho... |
136 |
static int page_outside_zone_boundaries(struct zone *zone, struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
137 |
{ |
bdc8cb984 [PATCH] memory ho... |
138 139 140 |
int ret = 0; unsigned seq; unsigned long pfn = page_to_pfn(page); |
c6a57e19e [PATCH] memory ho... |
141 |
|
bdc8cb984 [PATCH] memory ho... |
142 143 144 145 146 147 148 149 150 |
do { seq = zone_span_seqbegin(zone); if (pfn >= zone->zone_start_pfn + zone->spanned_pages) ret = 1; else if (pfn < zone->zone_start_pfn) ret = 1; } while (zone_span_seqretry(zone, seq)); return ret; |
c6a57e19e [PATCH] memory ho... |
151 152 153 154 |
} static int page_is_consistent(struct zone *zone, struct page *page) { |
1da177e4c Linux-2.6.12-rc2 |
155 156 |
#ifdef CONFIG_HOLES_IN_ZONE if (!pfn_valid(page_to_pfn(page))) |
c6a57e19e [PATCH] memory ho... |
157 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
158 159 |
#endif if (zone != page_zone(page)) |
c6a57e19e [PATCH] memory ho... |
160 161 162 163 164 165 166 167 168 169 |
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 |
170 |
return 1; |
c6a57e19e [PATCH] memory ho... |
171 172 |
if (!page_is_consistent(zone, page)) return 1; |
1da177e4c Linux-2.6.12-rc2 |
173 174 |
return 0; } |
13e7444b0 [PATCH] mm: remov... |
175 176 177 178 179 180 |
#else static inline int bad_range(struct zone *zone, struct page *page) { return 0; } #endif |
224abf92b [PATCH] mm: bad_p... |
181 |
static void bad_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
182 |
{ |
224abf92b [PATCH] mm: bad_p... |
183 184 |
printk(KERN_EMERG "Bad page state in process '%s' " |
7365f3d16 [PATCH] Restore K... |
185 186 187 188 189 190 |
KERN_EMERG "page:%p flags:0x%0*lx mapping:%p mapcount:%d count:%d " KERN_EMERG "Trying to fix it up, but a reboot is needed " KERN_EMERG "Backtrace: ", |
224abf92b [PATCH] mm: bad_p... |
191 192 193 |
current->comm, page, (int)(2*sizeof(unsigned long)), (unsigned long)page->flags, page->mapping, page_mapcount(page), page_count(page)); |
1da177e4c Linux-2.6.12-rc2 |
194 |
dump_stack(); |
334795eca [PATCH] bad_page:... |
195 196 |
page->flags &= ~(1 << PG_lru | 1 << PG_private | |
1da177e4c Linux-2.6.12-rc2 |
197 |
1 << PG_locked | |
1da177e4c Linux-2.6.12-rc2 |
198 199 |
1 << PG_active | 1 << PG_dirty | |
334795eca [PATCH] bad_page:... |
200 201 |
1 << PG_reclaim | 1 << PG_slab | |
1da177e4c Linux-2.6.12-rc2 |
202 |
1 << PG_swapcache | |
676165a8a [PATCH] Fix buddy... |
203 204 |
1 << PG_writeback | 1 << PG_buddy ); |
1da177e4c Linux-2.6.12-rc2 |
205 206 207 |
set_page_count(page, 0); reset_page_mapcount(page); page->mapping = NULL; |
9f1583339 [PATCH] use add_t... |
208 |
add_taint(TAINT_BAD_PAGE); |
1da177e4c Linux-2.6.12-rc2 |
209 |
} |
1da177e4c Linux-2.6.12-rc2 |
210 211 212 213 214 215 216 217 218 219 |
/* * 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". * * All pages have PG_compound set. All pages have their ->private pointing at * the head page (even the head page has this). * |
41d78ba55 [PATCH] compound ... |
220 221 222 |
* 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 |
223 |
*/ |
d98c7a098 [PATCH] compound ... |
224 225 226 227 228 |
static void free_compound_page(struct page *page) { __free_pages_ok(page, (unsigned long)page[1].lru.prev); } |
1da177e4c Linux-2.6.12-rc2 |
229 230 231 232 |
static void prep_compound_page(struct page *page, unsigned long order) { int i; int nr_pages = 1 << order; |
d98c7a098 [PATCH] compound ... |
233 |
page[1].lru.next = (void *)free_compound_page; /* set dtor */ |
41d78ba55 [PATCH] compound ... |
234 |
page[1].lru.prev = (void *)order; |
1da177e4c Linux-2.6.12-rc2 |
235 236 |
for (i = 0; i < nr_pages; i++) { struct page *p = page + i; |
5e9dace8d [PATCH] mm: page_... |
237 |
__SetPageCompound(p); |
4c21e2f24 [PATCH] mm: split... |
238 |
set_page_private(p, (unsigned long)page); |
1da177e4c Linux-2.6.12-rc2 |
239 240 241 242 243 244 245 |
} } static void destroy_compound_page(struct page *page, unsigned long order) { int i; int nr_pages = 1 << order; |
41d78ba55 [PATCH] compound ... |
246 |
if (unlikely((unsigned long)page[1].lru.prev != order)) |
224abf92b [PATCH] mm: bad_p... |
247 |
bad_page(page); |
1da177e4c Linux-2.6.12-rc2 |
248 249 250 |
for (i = 0; i < nr_pages; i++) { struct page *p = page + i; |
224abf92b [PATCH] mm: bad_p... |
251 252 253 |
if (unlikely(!PageCompound(p) | (page_private(p) != (unsigned long)page))) bad_page(page); |
5e9dace8d [PATCH] mm: page_... |
254 |
__ClearPageCompound(p); |
1da177e4c Linux-2.6.12-rc2 |
255 256 |
} } |
1da177e4c Linux-2.6.12-rc2 |
257 |
|
17cf44064 [PATCH] mm: clean... |
258 259 260 |
static inline void prep_zero_page(struct page *page, int order, gfp_t gfp_flags) { int i; |
725d704ec [PATCH] mm: VM_BU... |
261 |
VM_BUG_ON((gfp_flags & (__GFP_WAIT | __GFP_HIGHMEM)) == __GFP_HIGHMEM); |
6626c5d53 [PATCH] mm: prep_... |
262 263 264 265 |
/* * 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... |
266 |
VM_BUG_ON((gfp_flags & __GFP_HIGHMEM) && in_interrupt()); |
17cf44064 [PATCH] mm: clean... |
267 268 269 |
for (i = 0; i < (1 << order); i++) clear_highpage(page + i); } |
1da177e4c Linux-2.6.12-rc2 |
270 271 272 273 274 |
/* * function for dealing with page's order in buddy system. * zone->lock is already acquired when we use these. * So, we don't need atomic page->flags operations here. */ |
6aa3001b2 [PATCH] page_allo... |
275 276 |
static inline unsigned long page_order(struct page *page) { |
4c21e2f24 [PATCH] mm: split... |
277 |
return page_private(page); |
1da177e4c Linux-2.6.12-rc2 |
278 |
} |
6aa3001b2 [PATCH] page_allo... |
279 280 |
static inline void set_page_order(struct page *page, int order) { |
4c21e2f24 [PATCH] mm: split... |
281 |
set_page_private(page, order); |
676165a8a [PATCH] Fix buddy... |
282 |
__SetPageBuddy(page); |
1da177e4c Linux-2.6.12-rc2 |
283 284 285 286 |
} static inline void rmv_page_order(struct page *page) { |
676165a8a [PATCH] Fix buddy... |
287 |
__ClearPageBuddy(page); |
4c21e2f24 [PATCH] mm: split... |
288 |
set_page_private(page, 0); |
1da177e4c Linux-2.6.12-rc2 |
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 |
} /* * 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 |
306 |
* Assumption: *_mem_map is contiguous at least up to MAX_ORDER |
1da177e4c Linux-2.6.12-rc2 |
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 |
*/ static inline struct page * __page_find_buddy(struct page *page, unsigned long page_idx, unsigned int order) { unsigned long buddy_idx = page_idx ^ (1 << order); return page + (buddy_idx - page_idx); } static inline unsigned long __find_combined_index(unsigned long page_idx, unsigned int order) { return (page_idx & ~(1 << order)); } /* * 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... |
325 |
* (a) the buddy is not in a hole && |
676165a8a [PATCH] Fix buddy... |
326 |
* (b) the buddy is in the buddy system && |
cb2b95e1c [PATCH] zone hand... |
327 328 |
* (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... |
329 330 331 |
* * For recording whether a page is in the buddy system, we use PG_buddy. * Setting, clearing, and testing PG_buddy is serialized by zone->lock. |
1da177e4c Linux-2.6.12-rc2 |
332 |
* |
676165a8a [PATCH] Fix buddy... |
333 |
* For recording page's order, we use page_private(page). |
1da177e4c Linux-2.6.12-rc2 |
334 |
*/ |
cb2b95e1c [PATCH] zone hand... |
335 336 |
static inline int page_is_buddy(struct page *page, struct page *buddy, int order) |
1da177e4c Linux-2.6.12-rc2 |
337 |
{ |
13e7444b0 [PATCH] mm: remov... |
338 |
#ifdef CONFIG_HOLES_IN_ZONE |
cb2b95e1c [PATCH] zone hand... |
339 |
if (!pfn_valid(page_to_pfn(buddy))) |
13e7444b0 [PATCH] mm: remov... |
340 341 |
return 0; #endif |
cb2b95e1c [PATCH] zone hand... |
342 343 344 345 346 |
if (page_zone_id(page) != page_zone_id(buddy)) return 0; if (PageBuddy(buddy) && page_order(buddy) == order) { BUG_ON(page_count(buddy) != 0); |
6aa3001b2 [PATCH] page_allo... |
347 |
return 1; |
676165a8a [PATCH] Fix buddy... |
348 |
} |
6aa3001b2 [PATCH] page_allo... |
349 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 |
} /* * 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 |
676165a8a [PATCH] Fix buddy... |
365 |
* free pages of length of (1 << order) and marked with PG_buddy. Page's |
4c21e2f24 [PATCH] mm: split... |
366 |
* order is recorded in page_private(page) field. |
1da177e4c Linux-2.6.12-rc2 |
367 368 369 370 371 372 373 374 |
* So when we are allocating or freeing one, we can derive the state of the * other. That is, if we allocate a small block, and both were * free, the remainder of the region must be split into blocks. * If a block is freed, and its buddy is also free, then this * triggers coalescing into a block of larger size. * * -- wli */ |
48db57f8f [PATCH] mm: free_... |
375 |
static inline void __free_one_page(struct page *page, |
1da177e4c Linux-2.6.12-rc2 |
376 377 378 379 |
struct zone *zone, unsigned int order) { unsigned long page_idx; int order_size = 1 << order; |
224abf92b [PATCH] mm: bad_p... |
380 |
if (unlikely(PageCompound(page))) |
1da177e4c Linux-2.6.12-rc2 |
381 382 383 |
destroy_compound_page(page, order); page_idx = page_to_pfn(page) & ((1 << MAX_ORDER) - 1); |
725d704ec [PATCH] mm: VM_BU... |
384 385 |
VM_BUG_ON(page_idx & (order_size - 1)); VM_BUG_ON(bad_range(zone, page)); |
1da177e4c Linux-2.6.12-rc2 |
386 387 388 389 390 391 |
zone->free_pages += order_size; while (order < MAX_ORDER-1) { unsigned long combined_idx; struct free_area *area; struct page *buddy; |
1da177e4c Linux-2.6.12-rc2 |
392 |
buddy = __page_find_buddy(page, page_idx, order); |
cb2b95e1c [PATCH] zone hand... |
393 |
if (!page_is_buddy(page, buddy, order)) |
1da177e4c Linux-2.6.12-rc2 |
394 |
break; /* Move the buddy up one level. */ |
13e7444b0 [PATCH] mm: remov... |
395 |
|
1da177e4c Linux-2.6.12-rc2 |
396 397 398 399 |
list_del(&buddy->lru); area = zone->free_area + order; area->nr_free--; rmv_page_order(buddy); |
13e7444b0 [PATCH] mm: remov... |
400 |
combined_idx = __find_combined_index(page_idx, order); |
1da177e4c Linux-2.6.12-rc2 |
401 402 403 404 405 406 407 408 |
page = page + (combined_idx - page_idx); page_idx = combined_idx; order++; } set_page_order(page, order); list_add(&page->lru, &zone->free_area[order].free_list); zone->free_area[order].nr_free++; } |
224abf92b [PATCH] mm: bad_p... |
409 |
static inline int free_pages_check(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
410 |
{ |
92be2e33b [PATCH] mm: micro... |
411 412 413 |
if (unlikely(page_mapcount(page) | (page->mapping != NULL) | (page_count(page) != 0) | |
1da177e4c Linux-2.6.12-rc2 |
414 415 416 417 418 419 420 421 |
(page->flags & ( 1 << PG_lru | 1 << PG_private | 1 << PG_locked | 1 << PG_active | 1 << PG_reclaim | 1 << PG_slab | 1 << PG_swapcache | |
b5810039a [PATCH] core remo... |
422 |
1 << PG_writeback | |
676165a8a [PATCH] Fix buddy... |
423 424 |
1 << PG_reserved | 1 << PG_buddy )))) |
224abf92b [PATCH] mm: bad_p... |
425 |
bad_page(page); |
1da177e4c Linux-2.6.12-rc2 |
426 |
if (PageDirty(page)) |
242e54686 [PATCH] mm: remov... |
427 |
__ClearPageDirty(page); |
689bcebfd [PATCH] unpaged: ... |
428 429 430 431 432 433 |
/* * For now, we report if PG_reserved was found set, but do not * clear it, and do not free the page. But we shall soon need * to do more, for when the ZERO_PAGE count wraps negative. */ return PageReserved(page); |
1da177e4c Linux-2.6.12-rc2 |
434 435 436 437 438 |
} /* * Frees a list of pages. * Assumes all pages on list are in same zone, and of same order. |
207f36eec [PATCH] remove in... |
439 |
* count is the number of pages to free. |
1da177e4c Linux-2.6.12-rc2 |
440 441 442 443 444 445 446 |
* * 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. */ |
48db57f8f [PATCH] mm: free_... |
447 448 |
static void free_pages_bulk(struct zone *zone, int count, struct list_head *list, int order) |
1da177e4c Linux-2.6.12-rc2 |
449 |
{ |
c54ad30c7 [PATCH] mm: pagea... |
450 |
spin_lock(&zone->lock); |
1da177e4c Linux-2.6.12-rc2 |
451 452 |
zone->all_unreclaimable = 0; zone->pages_scanned = 0; |
48db57f8f [PATCH] mm: free_... |
453 454 |
while (count--) { struct page *page; |
725d704ec [PATCH] mm: VM_BU... |
455 |
VM_BUG_ON(list_empty(list)); |
1da177e4c Linux-2.6.12-rc2 |
456 |
page = list_entry(list->prev, struct page, lru); |
48db57f8f [PATCH] mm: free_... |
457 |
/* have to delete it as __free_one_page list manipulates */ |
1da177e4c Linux-2.6.12-rc2 |
458 |
list_del(&page->lru); |
48db57f8f [PATCH] mm: free_... |
459 |
__free_one_page(page, zone, order); |
1da177e4c Linux-2.6.12-rc2 |
460 |
} |
c54ad30c7 [PATCH] mm: pagea... |
461 |
spin_unlock(&zone->lock); |
1da177e4c Linux-2.6.12-rc2 |
462 |
} |
48db57f8f [PATCH] mm: free_... |
463 |
static void free_one_page(struct zone *zone, struct page *page, int order) |
1da177e4c Linux-2.6.12-rc2 |
464 |
{ |
006d22d9b [PATCH] Optimize ... |
465 466 467 468 469 |
spin_lock(&zone->lock); zone->all_unreclaimable = 0; zone->pages_scanned = 0; __free_one_page(page, zone ,order); spin_unlock(&zone->lock); |
48db57f8f [PATCH] mm: free_... |
470 471 472 473 474 |
} static void __free_pages_ok(struct page *page, unsigned int order) { unsigned long flags; |
1da177e4c Linux-2.6.12-rc2 |
475 |
int i; |
689bcebfd [PATCH] unpaged: ... |
476 |
int reserved = 0; |
1da177e4c Linux-2.6.12-rc2 |
477 |
|
1da177e4c Linux-2.6.12-rc2 |
478 |
for (i = 0 ; i < (1 << order) ; ++i) |
224abf92b [PATCH] mm: bad_p... |
479 |
reserved += free_pages_check(page + i); |
689bcebfd [PATCH] unpaged: ... |
480 481 |
if (reserved) return; |
9858db504 [PATCH] mm: locks... |
482 483 |
if (!PageHighMem(page)) debug_check_no_locks_freed(page_address(page),PAGE_SIZE<<order); |
dafb13673 [PATCH] mm: arch_... |
484 |
arch_free_page(page, order); |
48db57f8f [PATCH] mm: free_... |
485 |
kernel_map_pages(page, 1 << order, 0); |
dafb13673 [PATCH] mm: arch_... |
486 |
|
c54ad30c7 [PATCH] mm: pagea... |
487 |
local_irq_save(flags); |
f8891e5e1 [PATCH] Light wei... |
488 |
__count_vm_events(PGFREE, 1 << order); |
48db57f8f [PATCH] mm: free_... |
489 |
free_one_page(page_zone(page), page, order); |
c54ad30c7 [PATCH] mm: pagea... |
490 |
local_irq_restore(flags); |
1da177e4c Linux-2.6.12-rc2 |
491 |
} |
a226f6c89 [PATCH] FRV: Clea... |
492 493 494 495 496 497 498 499 |
/* * permit the bootmem allocator to evade page validation on high-order frees */ void fastcall __init __free_pages_bootmem(struct page *page, unsigned int order) { if (order == 0) { __ClearPageReserved(page); set_page_count(page, 0); |
7835e98b2 [PATCH] remove se... |
500 |
set_page_refcounted(page); |
545b1ea9b [PATCH] mm: clean... |
501 |
__free_page(page); |
a226f6c89 [PATCH] FRV: Clea... |
502 |
} else { |
a226f6c89 [PATCH] FRV: Clea... |
503 |
int loop; |
545b1ea9b [PATCH] mm: clean... |
504 |
prefetchw(page); |
a226f6c89 [PATCH] FRV: Clea... |
505 506 |
for (loop = 0; loop < BITS_PER_LONG; loop++) { struct page *p = &page[loop]; |
545b1ea9b [PATCH] mm: clean... |
507 508 |
if (loop + 1 < BITS_PER_LONG) prefetchw(p + 1); |
a226f6c89 [PATCH] FRV: Clea... |
509 510 511 |
__ClearPageReserved(p); set_page_count(p, 0); } |
7835e98b2 [PATCH] remove se... |
512 |
set_page_refcounted(page); |
545b1ea9b [PATCH] mm: clean... |
513 |
__free_pages(page, order); |
a226f6c89 [PATCH] FRV: Clea... |
514 515 |
} } |
1da177e4c Linux-2.6.12-rc2 |
516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 |
/* * 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. * * -- wli */ |
085cc7d5d [PATCH] mm: page_... |
531 |
static inline void expand(struct zone *zone, struct page *page, |
1da177e4c Linux-2.6.12-rc2 |
532 533 534 535 536 537 538 539 |
int low, int high, struct free_area *area) { unsigned long size = 1 << high; while (high > low) { area--; high--; size >>= 1; |
725d704ec [PATCH] mm: VM_BU... |
540 |
VM_BUG_ON(bad_range(zone, &page[size])); |
1da177e4c Linux-2.6.12-rc2 |
541 542 543 544 |
list_add(&page[size].lru, &area->free_list); area->nr_free++; set_page_order(&page[size], high); } |
1da177e4c Linux-2.6.12-rc2 |
545 |
} |
1da177e4c Linux-2.6.12-rc2 |
546 547 548 |
/* * This page is about to be returned from the page allocator */ |
17cf44064 [PATCH] mm: clean... |
549 |
static int prep_new_page(struct page *page, int order, gfp_t gfp_flags) |
1da177e4c Linux-2.6.12-rc2 |
550 |
{ |
92be2e33b [PATCH] mm: micro... |
551 552 553 |
if (unlikely(page_mapcount(page) | (page->mapping != NULL) | (page_count(page) != 0) | |
334795eca [PATCH] bad_page:... |
554 555 |
(page->flags & ( 1 << PG_lru | |
1da177e4c Linux-2.6.12-rc2 |
556 557 |
1 << PG_private | 1 << PG_locked | |
1da177e4c Linux-2.6.12-rc2 |
558 559 560 |
1 << PG_active | 1 << PG_dirty | 1 << PG_reclaim | |
334795eca [PATCH] bad_page:... |
561 |
1 << PG_slab | |
1da177e4c Linux-2.6.12-rc2 |
562 |
1 << PG_swapcache | |
b5810039a [PATCH] core remo... |
563 |
1 << PG_writeback | |
676165a8a [PATCH] Fix buddy... |
564 565 |
1 << PG_reserved | 1 << PG_buddy )))) |
224abf92b [PATCH] mm: bad_p... |
566 |
bad_page(page); |
1da177e4c Linux-2.6.12-rc2 |
567 |
|
689bcebfd [PATCH] unpaged: ... |
568 569 570 571 572 573 |
/* * For now, we report if PG_reserved was found set, but do not * clear it, and do not allocate the page: as a safety net. */ if (PageReserved(page)) return 1; |
1da177e4c Linux-2.6.12-rc2 |
574 575 576 |
page->flags &= ~(1 << PG_uptodate | 1 << PG_error | 1 << PG_referenced | 1 << PG_arch_1 | 1 << PG_checked | 1 << PG_mappedtodisk); |
4c21e2f24 [PATCH] mm: split... |
577 |
set_page_private(page, 0); |
7835e98b2 [PATCH] remove se... |
578 |
set_page_refcounted(page); |
1da177e4c Linux-2.6.12-rc2 |
579 |
kernel_map_pages(page, 1 << order, 1); |
17cf44064 [PATCH] mm: clean... |
580 581 582 583 584 585 |
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: ... |
586 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 |
} /* * Do the hard work of removing an element from the buddy allocator. * Call me with the zone->lock already held. */ static struct page *__rmqueue(struct zone *zone, unsigned int order) { struct free_area * area; unsigned int current_order; struct page *page; for (current_order = order; current_order < MAX_ORDER; ++current_order) { area = zone->free_area + current_order; if (list_empty(&area->free_list)) continue; page = list_entry(area->free_list.next, struct page, lru); list_del(&page->lru); rmv_page_order(page); area->nr_free--; zone->free_pages -= 1UL << order; |
085cc7d5d [PATCH] mm: page_... |
609 610 |
expand(zone, page, order, current_order, area); return page; |
1da177e4c Linux-2.6.12-rc2 |
611 612 613 614 615 616 617 618 619 620 621 622 623 |
} return NULL; } /* * 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. */ static int rmqueue_bulk(struct zone *zone, unsigned int order, unsigned long count, struct list_head *list) { |
1da177e4c Linux-2.6.12-rc2 |
624 |
int i; |
1da177e4c Linux-2.6.12-rc2 |
625 |
|
c54ad30c7 [PATCH] mm: pagea... |
626 |
spin_lock(&zone->lock); |
1da177e4c Linux-2.6.12-rc2 |
627 |
for (i = 0; i < count; ++i) { |
085cc7d5d [PATCH] mm: page_... |
628 629 |
struct page *page = __rmqueue(zone, order); if (unlikely(page == NULL)) |
1da177e4c Linux-2.6.12-rc2 |
630 |
break; |
1da177e4c Linux-2.6.12-rc2 |
631 632 |
list_add_tail(&page->lru, list); } |
c54ad30c7 [PATCH] mm: pagea... |
633 |
spin_unlock(&zone->lock); |
085cc7d5d [PATCH] mm: page_... |
634 |
return i; |
1da177e4c Linux-2.6.12-rc2 |
635 |
} |
4ae7c0394 [PATCH] Periodica... |
636 |
#ifdef CONFIG_NUMA |
8fce4d8e3 [PATCH] slab: Nod... |
637 638 |
/* * Called from the slab reaper to drain pagesets on a particular node that |
39bbcb8f8 [PATCH] mm: do no... |
639 |
* belongs to the currently executing processor. |
879336c39 [PATCH] drain_nod... |
640 641 |
* Note that this function must be called with the thread pinned to * a single processor. |
8fce4d8e3 [PATCH] slab: Nod... |
642 643 |
*/ void drain_node_pages(int nodeid) |
4ae7c0394 [PATCH] Periodica... |
644 |
{ |
2f6726e54 [PATCH] Apply typ... |
645 646 |
int i; enum zone_type z; |
4ae7c0394 [PATCH] Periodica... |
647 |
unsigned long flags; |
8fce4d8e3 [PATCH] slab: Nod... |
648 649 |
for (z = 0; z < MAX_NR_ZONES; z++) { struct zone *zone = NODE_DATA(nodeid)->node_zones + z; |
4ae7c0394 [PATCH] Periodica... |
650 |
struct per_cpu_pageset *pset; |
39bbcb8f8 [PATCH] mm: do no... |
651 652 |
if (!populated_zone(zone)) continue; |
23316bc86 [PATCH] mm: clean... |
653 |
pset = zone_pcp(zone, smp_processor_id()); |
4ae7c0394 [PATCH] Periodica... |
654 655 656 657 |
for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) { struct per_cpu_pages *pcp; pcp = &pset->pcp[i]; |
879336c39 [PATCH] drain_nod... |
658 659 660 661 662 663 |
if (pcp->count) { local_irq_save(flags); free_pages_bulk(zone, pcp->count, &pcp->list, 0); pcp->count = 0; local_irq_restore(flags); } |
4ae7c0394 [PATCH] Periodica... |
664 665 |
} } |
4ae7c0394 [PATCH] Periodica... |
666 667 |
} #endif |
1da177e4c Linux-2.6.12-rc2 |
668 669 670 |
#if defined(CONFIG_PM) || defined(CONFIG_HOTPLUG_CPU) static void __drain_pages(unsigned int cpu) { |
c54ad30c7 [PATCH] mm: pagea... |
671 |
unsigned long flags; |
1da177e4c Linux-2.6.12-rc2 |
672 673 674 675 676 |
struct zone *zone; int i; for_each_zone(zone) { struct per_cpu_pageset *pset; |
e7c8d5c99 [PATCH] node loca... |
677 |
pset = zone_pcp(zone, cpu); |
1da177e4c Linux-2.6.12-rc2 |
678 679 680 681 |
for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) { struct per_cpu_pages *pcp; pcp = &pset->pcp[i]; |
c54ad30c7 [PATCH] mm: pagea... |
682 |
local_irq_save(flags); |
48db57f8f [PATCH] mm: free_... |
683 684 |
free_pages_bulk(zone, pcp->count, &pcp->list, 0); pcp->count = 0; |
c54ad30c7 [PATCH] mm: pagea... |
685 |
local_irq_restore(flags); |
1da177e4c Linux-2.6.12-rc2 |
686 687 688 689 690 691 692 693 694 |
} } } #endif /* CONFIG_PM || CONFIG_HOTPLUG_CPU */ #ifdef CONFIG_PM void mark_free_pages(struct zone *zone) { |
f623f0db8 [PATCH] swsusp: F... |
695 696 |
unsigned long pfn, max_zone_pfn; unsigned long flags; |
1da177e4c Linux-2.6.12-rc2 |
697 698 699 700 701 702 703 |
int order; struct list_head *curr; if (!zone->spanned_pages) return; spin_lock_irqsave(&zone->lock, flags); |
f623f0db8 [PATCH] swsusp: F... |
704 705 706 707 708 709 710 711 712 |
max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) if (pfn_valid(pfn)) { struct page *page = pfn_to_page(pfn); if (!PageNosave(page)) ClearPageNosaveFree(page); } |
1da177e4c Linux-2.6.12-rc2 |
713 714 715 |
for (order = MAX_ORDER - 1; order >= 0; --order) list_for_each(curr, &zone->free_area[order].free_list) { |
f623f0db8 [PATCH] swsusp: F... |
716 |
unsigned long i; |
1da177e4c Linux-2.6.12-rc2 |
717 |
|
f623f0db8 [PATCH] swsusp: F... |
718 719 720 721 |
pfn = page_to_pfn(list_entry(curr, struct page, lru)); for (i = 0; i < (1UL << order); i++) SetPageNosaveFree(pfn_to_page(pfn + i)); } |
1da177e4c Linux-2.6.12-rc2 |
722 |
|
1da177e4c Linux-2.6.12-rc2 |
723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 |
spin_unlock_irqrestore(&zone->lock, flags); } /* * Spill all of this CPU's per-cpu pages back into the buddy allocator. */ void drain_local_pages(void) { unsigned long flags; local_irq_save(flags); __drain_pages(smp_processor_id()); local_irq_restore(flags); } #endif /* CONFIG_PM */ |
1da177e4c Linux-2.6.12-rc2 |
738 739 740 |
/* * Free a 0-order page */ |
1da177e4c Linux-2.6.12-rc2 |
741 742 743 744 745 |
static void fastcall free_hot_cold_page(struct page *page, int cold) { struct zone *zone = page_zone(page); struct per_cpu_pages *pcp; unsigned long flags; |
1da177e4c Linux-2.6.12-rc2 |
746 747 |
if (PageAnon(page)) page->mapping = NULL; |
224abf92b [PATCH] mm: bad_p... |
748 |
if (free_pages_check(page)) |
689bcebfd [PATCH] unpaged: ... |
749 |
return; |
9858db504 [PATCH] mm: locks... |
750 751 |
if (!PageHighMem(page)) debug_check_no_locks_freed(page_address(page), PAGE_SIZE); |
dafb13673 [PATCH] mm: arch_... |
752 |
arch_free_page(page, 0); |
689bcebfd [PATCH] unpaged: ... |
753 |
kernel_map_pages(page, 1, 0); |
e7c8d5c99 [PATCH] node loca... |
754 |
pcp = &zone_pcp(zone, get_cpu())->pcp[cold]; |
1da177e4c Linux-2.6.12-rc2 |
755 |
local_irq_save(flags); |
f8891e5e1 [PATCH] Light wei... |
756 |
__count_vm_event(PGFREE); |
1da177e4c Linux-2.6.12-rc2 |
757 758 |
list_add(&page->lru, &pcp->list); pcp->count++; |
48db57f8f [PATCH] mm: free_... |
759 760 761 762 |
if (pcp->count >= pcp->high) { free_pages_bulk(zone, pcp->batch, &pcp->list, 0); pcp->count -= pcp->batch; } |
1da177e4c Linux-2.6.12-rc2 |
763 764 765 766 767 768 769 770 771 772 773 774 775 |
local_irq_restore(flags); put_cpu(); } void fastcall free_hot_page(struct page *page) { free_hot_cold_page(page, 0); } void fastcall free_cold_page(struct page *page) { free_hot_cold_page(page, 1); } |
8dfcc9ba2 [PATCH] mm: split... |
776 777 778 779 780 781 782 783 784 785 786 |
/* * 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... |
787 788 |
VM_BUG_ON(PageCompound(page)); VM_BUG_ON(!page_count(page)); |
7835e98b2 [PATCH] remove se... |
789 790 |
for (i = 1; i < (1 << order); i++) set_page_refcounted(page + i); |
8dfcc9ba2 [PATCH] mm: split... |
791 |
} |
8dfcc9ba2 [PATCH] mm: split... |
792 |
|
1da177e4c Linux-2.6.12-rc2 |
793 794 795 796 797 |
/* * 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. */ |
a74609faf [PATCH] mm: page_... |
798 799 |
static struct page *buffered_rmqueue(struct zonelist *zonelist, struct zone *zone, int order, gfp_t gfp_flags) |
1da177e4c Linux-2.6.12-rc2 |
800 801 |
{ unsigned long flags; |
689bcebfd [PATCH] unpaged: ... |
802 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
803 |
int cold = !!(gfp_flags & __GFP_COLD); |
a74609faf [PATCH] mm: page_... |
804 |
int cpu; |
1da177e4c Linux-2.6.12-rc2 |
805 |
|
689bcebfd [PATCH] unpaged: ... |
806 |
again: |
a74609faf [PATCH] mm: page_... |
807 |
cpu = get_cpu(); |
48db57f8f [PATCH] mm: free_... |
808 |
if (likely(order == 0)) { |
1da177e4c Linux-2.6.12-rc2 |
809 |
struct per_cpu_pages *pcp; |
a74609faf [PATCH] mm: page_... |
810 |
pcp = &zone_pcp(zone, cpu)->pcp[cold]; |
1da177e4c Linux-2.6.12-rc2 |
811 |
local_irq_save(flags); |
a74609faf [PATCH] mm: page_... |
812 |
if (!pcp->count) { |
1da177e4c Linux-2.6.12-rc2 |
813 814 |
pcp->count += rmqueue_bulk(zone, 0, pcp->batch, &pcp->list); |
a74609faf [PATCH] mm: page_... |
815 816 |
if (unlikely(!pcp->count)) goto failed; |
1da177e4c Linux-2.6.12-rc2 |
817 |
} |
a74609faf [PATCH] mm: page_... |
818 819 820 |
page = list_entry(pcp->list.next, struct page, lru); list_del(&page->lru); pcp->count--; |
7fb1d9fca [PATCH] mm: __all... |
821 |
} else { |
1da177e4c Linux-2.6.12-rc2 |
822 823 |
spin_lock_irqsave(&zone->lock, flags); page = __rmqueue(zone, order); |
a74609faf [PATCH] mm: page_... |
824 825 826 |
spin_unlock(&zone->lock); if (!page) goto failed; |
1da177e4c Linux-2.6.12-rc2 |
827 |
} |
f8891e5e1 [PATCH] Light wei... |
828 |
__count_zone_vm_events(PGALLOC, zone, 1 << order); |
ca889e6c4 [PATCH] Use Zoned... |
829 |
zone_statistics(zonelist, zone); |
a74609faf [PATCH] mm: page_... |
830 831 |
local_irq_restore(flags); put_cpu(); |
1da177e4c Linux-2.6.12-rc2 |
832 |
|
725d704ec [PATCH] mm: VM_BU... |
833 |
VM_BUG_ON(bad_range(zone, page)); |
17cf44064 [PATCH] mm: clean... |
834 |
if (prep_new_page(page, order, gfp_flags)) |
a74609faf [PATCH] mm: page_... |
835 |
goto again; |
1da177e4c Linux-2.6.12-rc2 |
836 |
return page; |
a74609faf [PATCH] mm: page_... |
837 838 839 840 841 |
failed: local_irq_restore(flags); put_cpu(); return NULL; |
1da177e4c Linux-2.6.12-rc2 |
842 |
} |
7fb1d9fca [PATCH] mm: __all... |
843 |
#define ALLOC_NO_WATERMARKS 0x01 /* don't check watermarks at all */ |
3148890bf [PATCH] mm: __all... |
844 845 846 847 848 849 |
#define ALLOC_WMARK_MIN 0x02 /* use pages_min watermark */ #define ALLOC_WMARK_LOW 0x04 /* use pages_low watermark */ #define ALLOC_WMARK_HIGH 0x08 /* use pages_high watermark */ #define ALLOC_HARDER 0x10 /* try to alloc harder */ #define ALLOC_HIGH 0x20 /* __GFP_HIGH set */ #define ALLOC_CPUSET 0x40 /* check for correct cpuset */ |
7fb1d9fca [PATCH] mm: __all... |
850 |
|
1da177e4c Linux-2.6.12-rc2 |
851 852 853 854 855 |
/* * Return 1 if free pages are above 'mark'. This takes into account the order * of the allocation. */ int zone_watermark_ok(struct zone *z, int order, unsigned long mark, |
7fb1d9fca [PATCH] mm: __all... |
856 |
int classzone_idx, int alloc_flags) |
1da177e4c Linux-2.6.12-rc2 |
857 858 |
{ /* free_pages my go negative - that's OK */ |
e80ee884a [PATCH] mm: micro... |
859 860 |
unsigned long min = mark; long free_pages = z->free_pages - (1 << order) + 1; |
1da177e4c Linux-2.6.12-rc2 |
861 |
int o; |
7fb1d9fca [PATCH] mm: __all... |
862 |
if (alloc_flags & ALLOC_HIGH) |
1da177e4c Linux-2.6.12-rc2 |
863 |
min -= min / 2; |
7fb1d9fca [PATCH] mm: __all... |
864 |
if (alloc_flags & ALLOC_HARDER) |
1da177e4c Linux-2.6.12-rc2 |
865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 |
min -= min / 4; if (free_pages <= min + z->lowmem_reserve[classzone_idx]) return 0; 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) return 0; } return 1; } |
7fb1d9fca [PATCH] mm: __all... |
881 882 883 884 885 886 887 |
/* * get_page_from_freeliest goes through the zonelist trying to allocate * a page. */ static struct page * get_page_from_freelist(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, int alloc_flags) |
753ee7289 [PATCH] VM: early... |
888 |
{ |
7fb1d9fca [PATCH] mm: __all... |
889 890 891 |
struct zone **z = zonelist->zones; struct page *page = NULL; int classzone_idx = zone_idx(*z); |
1192d5264 [PATCH] Cleanup: ... |
892 |
struct zone *zone; |
7fb1d9fca [PATCH] mm: __all... |
893 894 895 896 897 898 |
/* * Go through the zonelist once, looking for a zone with enough free. * See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ do { |
1192d5264 [PATCH] Cleanup: ... |
899 |
zone = *z; |
08e0f6a97 [PATCH] Add NUMA_... |
900 |
if (unlikely(NUMA_BUILD && (gfp_mask & __GFP_THISNODE) && |
1192d5264 [PATCH] Cleanup: ... |
901 |
zone->zone_pgdat != zonelist->zones[0]->zone_pgdat)) |
9b819d204 [PATCH] Add __GFP... |
902 |
break; |
7fb1d9fca [PATCH] mm: __all... |
903 |
if ((alloc_flags & ALLOC_CPUSET) && |
1192d5264 [PATCH] Cleanup: ... |
904 |
!cpuset_zone_allowed(zone, gfp_mask)) |
7fb1d9fca [PATCH] mm: __all... |
905 906 907 |
continue; if (!(alloc_flags & ALLOC_NO_WATERMARKS)) { |
3148890bf [PATCH] mm: __all... |
908 909 |
unsigned long mark; if (alloc_flags & ALLOC_WMARK_MIN) |
1192d5264 [PATCH] Cleanup: ... |
910 |
mark = zone->pages_min; |
3148890bf [PATCH] mm: __all... |
911 |
else if (alloc_flags & ALLOC_WMARK_LOW) |
1192d5264 [PATCH] Cleanup: ... |
912 |
mark = zone->pages_low; |
3148890bf [PATCH] mm: __all... |
913 |
else |
1192d5264 [PATCH] Cleanup: ... |
914 915 |
mark = zone->pages_high; if (!zone_watermark_ok(zone , order, mark, |
7fb1d9fca [PATCH] mm: __all... |
916 |
classzone_idx, alloc_flags)) |
9eeff2395 [PATCH] Zone recl... |
917 |
if (!zone_reclaim_mode || |
1192d5264 [PATCH] Cleanup: ... |
918 |
!zone_reclaim(zone, gfp_mask, order)) |
9eeff2395 [PATCH] Zone recl... |
919 |
continue; |
7fb1d9fca [PATCH] mm: __all... |
920 |
} |
1192d5264 [PATCH] Cleanup: ... |
921 |
page = buffered_rmqueue(zonelist, zone, order, gfp_mask); |
7fb1d9fca [PATCH] mm: __all... |
922 |
if (page) { |
7fb1d9fca [PATCH] mm: __all... |
923 924 925 926 |
break; } } while (*(++z) != NULL); return page; |
753ee7289 [PATCH] VM: early... |
927 |
} |
1da177e4c Linux-2.6.12-rc2 |
928 929 930 931 |
/* * This is the 'heart' of the zoned buddy allocator. */ struct page * fastcall |
dd0fc66fb [PATCH] gfp flags... |
932 |
__alloc_pages(gfp_t gfp_mask, unsigned int order, |
1da177e4c Linux-2.6.12-rc2 |
933 934 |
struct zonelist *zonelist) { |
260b23674 [PATCH] gfp_t: th... |
935 |
const gfp_t wait = gfp_mask & __GFP_WAIT; |
7fb1d9fca [PATCH] mm: __all... |
936 |
struct zone **z; |
1da177e4c Linux-2.6.12-rc2 |
937 938 939 |
struct page *page; struct reclaim_state reclaim_state; struct task_struct *p = current; |
1da177e4c Linux-2.6.12-rc2 |
940 |
int do_retry; |
7fb1d9fca [PATCH] mm: __all... |
941 |
int alloc_flags; |
1da177e4c Linux-2.6.12-rc2 |
942 943 944 |
int did_some_progress; might_sleep_if(wait); |
6b1de9161 [PATCH] VM: fix z... |
945 |
restart: |
7fb1d9fca [PATCH] mm: __all... |
946 |
z = zonelist->zones; /* the list of zones suitable for gfp_mask */ |
1da177e4c Linux-2.6.12-rc2 |
947 |
|
7fb1d9fca [PATCH] mm: __all... |
948 |
if (unlikely(*z == NULL)) { |
1da177e4c Linux-2.6.12-rc2 |
949 950 951 |
/* Should this ever happen?? */ return NULL; } |
6b1de9161 [PATCH] VM: fix z... |
952 |
|
7fb1d9fca [PATCH] mm: __all... |
953 |
page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, order, |
3148890bf [PATCH] mm: __all... |
954 |
zonelist, ALLOC_WMARK_LOW|ALLOC_CPUSET); |
7fb1d9fca [PATCH] mm: __all... |
955 956 |
if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
957 |
|
6b1de9161 [PATCH] VM: fix z... |
958 |
do { |
43b0bc00f [PATCH] cpuset: r... |
959 |
wakeup_kswapd(*z, order); |
6b1de9161 [PATCH] VM: fix z... |
960 |
} while (*(++z)); |
1da177e4c Linux-2.6.12-rc2 |
961 |
|
9bf2229f8 [PATCH] cpusets: ... |
962 |
/* |
7fb1d9fca [PATCH] mm: __all... |
963 964 965 966 967 968 |
* 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. * * The caller may dip into page reserves a bit more if the caller * cannot run direct reclaim, or if the caller has realtime scheduling |
4eac915d0 [PATCH] mm: gfp_a... |
969 970 |
* policy or is asking for __GFP_HIGH memory. GFP_ATOMIC requests will * set both ALLOC_HARDER (!wait) and ALLOC_HIGH (__GFP_HIGH). |
9bf2229f8 [PATCH] cpusets: ... |
971 |
*/ |
3148890bf [PATCH] mm: __all... |
972 |
alloc_flags = ALLOC_WMARK_MIN; |
7fb1d9fca [PATCH] mm: __all... |
973 974 975 976 |
if ((unlikely(rt_task(p)) && !in_interrupt()) || !wait) alloc_flags |= ALLOC_HARDER; if (gfp_mask & __GFP_HIGH) alloc_flags |= ALLOC_HIGH; |
bdd804f47 [PATCH] Cpuset: m... |
977 978 |
if (wait) alloc_flags |= ALLOC_CPUSET; |
1da177e4c Linux-2.6.12-rc2 |
979 980 981 |
/* * Go through the zonelist again. Let __GFP_HIGH and allocations |
7fb1d9fca [PATCH] mm: __all... |
982 |
* coming from realtime tasks go deeper into reserves. |
1da177e4c Linux-2.6.12-rc2 |
983 984 985 |
* * This is the last chance, in general, before the goto nopage. * Ignore cpuset if GFP_ATOMIC (!wait) rather than fail alloc. |
9bf2229f8 [PATCH] cpusets: ... |
986 |
* See also cpuset_zone_allowed() comment in kernel/cpuset.c. |
1da177e4c Linux-2.6.12-rc2 |
987 |
*/ |
7fb1d9fca [PATCH] mm: __all... |
988 989 990 |
page = get_page_from_freelist(gfp_mask, order, zonelist, alloc_flags); if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
991 992 |
/* This allocation should allow future memory freeing. */ |
b84a35be0 [PATCH] mempool: ... |
993 994 995 996 |
if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE))) && !in_interrupt()) { if (!(gfp_mask & __GFP_NOMEMALLOC)) { |
885036d32 [PATCH] mm: __GFP... |
997 |
nofail_alloc: |
b84a35be0 [PATCH] mempool: ... |
998 |
/* go through the zonelist yet again, ignoring mins */ |
7fb1d9fca [PATCH] mm: __all... |
999 |
page = get_page_from_freelist(gfp_mask, order, |
47f3a867f [PATCH] mm: fix _... |
1000 |
zonelist, ALLOC_NO_WATERMARKS); |
7fb1d9fca [PATCH] mm: __all... |
1001 1002 |
if (page) goto got_pg; |
885036d32 [PATCH] mm: __GFP... |
1003 1004 1005 1006 |
if (gfp_mask & __GFP_NOFAIL) { blk_congestion_wait(WRITE, HZ/50); goto nofail_alloc; } |
1da177e4c Linux-2.6.12-rc2 |
1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 |
} goto nopage; } /* Atomic allocations - we can't balance anything */ if (!wait) goto nopage; rebalance: cond_resched(); /* We now go into synchronous reclaim */ |
3e0d98b9f [PATCH] cpuset: m... |
1019 |
cpuset_memory_pressure_bump(); |
1da177e4c Linux-2.6.12-rc2 |
1020 1021 1022 |
p->flags |= PF_MEMALLOC; reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; |
7fb1d9fca [PATCH] mm: __all... |
1023 |
did_some_progress = try_to_free_pages(zonelist->zones, gfp_mask); |
1da177e4c Linux-2.6.12-rc2 |
1024 1025 1026 1027 1028 1029 1030 |
p->reclaim_state = NULL; p->flags &= ~PF_MEMALLOC; cond_resched(); if (likely(did_some_progress)) { |
7fb1d9fca [PATCH] mm: __all... |
1031 1032 1033 1034 |
page = get_page_from_freelist(gfp_mask, order, zonelist, alloc_flags); if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
1035 1036 1037 1038 1039 1040 1041 |
} else if ((gfp_mask & __GFP_FS) && !(gfp_mask & __GFP_NORETRY)) { /* * 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. */ |
7fb1d9fca [PATCH] mm: __all... |
1042 |
page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, order, |
3148890bf [PATCH] mm: __all... |
1043 |
zonelist, ALLOC_WMARK_HIGH|ALLOC_CPUSET); |
7fb1d9fca [PATCH] mm: __all... |
1044 1045 |
if (page) goto got_pg; |
1da177e4c Linux-2.6.12-rc2 |
1046 |
|
9b0f8b040 [PATCH] Terminate... |
1047 |
out_of_memory(zonelist, gfp_mask, order); |
1da177e4c Linux-2.6.12-rc2 |
1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 |
goto restart; } /* * Don't let big-order allocations loop unless the caller explicitly * requests that. Wait for some write requests to complete then retry. * * In this implementation, __GFP_REPEAT means __GFP_NOFAIL for order * <= 3, but that may not be true in other implementations. */ do_retry = 0; if (!(gfp_mask & __GFP_NORETRY)) { if ((order <= 3) || (gfp_mask & __GFP_REPEAT)) do_retry = 1; if (gfp_mask & __GFP_NOFAIL) do_retry = 1; } if (do_retry) { blk_congestion_wait(WRITE, HZ/50); goto rebalance; } nopage: if (!(gfp_mask & __GFP_NOWARN) && printk_ratelimit()) { printk(KERN_WARNING "%s: page allocation failure." " order:%d, mode:0x%x ", p->comm, order, gfp_mask); dump_stack(); |
578c2fd6a [PATCH] add OOM d... |
1077 |
show_mem(); |
1da177e4c Linux-2.6.12-rc2 |
1078 |
} |
1da177e4c Linux-2.6.12-rc2 |
1079 |
got_pg: |
1da177e4c Linux-2.6.12-rc2 |
1080 1081 1082 1083 1084 1085 1086 1087 |
return page; } EXPORT_SYMBOL(__alloc_pages); /* * Common helper functions. */ |
dd0fc66fb [PATCH] gfp flags... |
1088 |
fastcall unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order) |
1da177e4c Linux-2.6.12-rc2 |
1089 1090 1091 1092 1093 1094 1095 1096 1097 |
{ struct page * page; page = alloc_pages(gfp_mask, order); if (!page) return 0; return (unsigned long) page_address(page); } EXPORT_SYMBOL(__get_free_pages); |
dd0fc66fb [PATCH] gfp flags... |
1098 |
fastcall unsigned long get_zeroed_page(gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
1099 1100 1101 1102 1103 1104 1105 |
{ struct page * page; /* * get_zeroed_page() returns a 32-bit address, which cannot represent * a highmem page */ |
725d704ec [PATCH] mm: VM_BU... |
1106 |
VM_BUG_ON((gfp_mask & __GFP_HIGHMEM) != 0); |
1da177e4c Linux-2.6.12-rc2 |
1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 |
page = alloc_pages(gfp_mask | __GFP_ZERO, 0); if (page) return (unsigned long) page_address(page); return 0; } EXPORT_SYMBOL(get_zeroed_page); void __pagevec_free(struct pagevec *pvec) { int i = pagevec_count(pvec); while (--i >= 0) free_hot_cold_page(pvec->pages[i], pvec->cold); } fastcall void __free_pages(struct page *page, unsigned int order) { |
b5810039a [PATCH] core remo... |
1126 |
if (put_page_testzero(page)) { |
1da177e4c Linux-2.6.12-rc2 |
1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 |
if (order == 0) free_hot_page(page); else __free_pages_ok(page, order); } } EXPORT_SYMBOL(__free_pages); fastcall void free_pages(unsigned long addr, unsigned int order) { if (addr != 0) { |
725d704ec [PATCH] mm: VM_BU... |
1139 |
VM_BUG_ON(!virt_addr_valid((void *)addr)); |
1da177e4c Linux-2.6.12-rc2 |
1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 |
__free_pages(virt_to_page((void *)addr), order); } } EXPORT_SYMBOL(free_pages); /* * Total amount of free (allocatable) RAM: */ unsigned int nr_free_pages(void) { unsigned int sum = 0; struct zone *zone; for_each_zone(zone) sum += zone->free_pages; return sum; } EXPORT_SYMBOL(nr_free_pages); #ifdef CONFIG_NUMA unsigned int nr_free_pages_pgdat(pg_data_t *pgdat) { |
2f6726e54 [PATCH] Apply typ... |
1165 1166 |
unsigned int sum = 0; enum zone_type i; |
1da177e4c Linux-2.6.12-rc2 |
1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 |
for (i = 0; i < MAX_NR_ZONES; i++) sum += pgdat->node_zones[i].free_pages; return sum; } #endif static unsigned int nr_free_zone_pages(int offset) { |
e310fd432 [PATCH] Fix NUMA ... |
1177 1178 |
/* Just pick one node, since fallback list is circular */ pg_data_t *pgdat = NODE_DATA(numa_node_id()); |
1da177e4c Linux-2.6.12-rc2 |
1179 |
unsigned int sum = 0; |
e310fd432 [PATCH] Fix NUMA ... |
1180 1181 1182 |
struct zonelist *zonelist = pgdat->node_zonelists + offset; struct zone **zonep = zonelist->zones; struct zone *zone; |
1da177e4c Linux-2.6.12-rc2 |
1183 |
|
e310fd432 [PATCH] Fix NUMA ... |
1184 1185 1186 1187 1188 |
for (zone = *zonep++; zone; zone = *zonep++) { unsigned long size = zone->present_pages; unsigned long high = zone->pages_high; if (size > high) sum += size - high; |
1da177e4c Linux-2.6.12-rc2 |
1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 |
} return sum; } /* * Amount of free RAM allocatable within ZONE_DMA and ZONE_NORMAL */ unsigned int nr_free_buffer_pages(void) { |
af4ca457e [PATCH] gfp_t: in... |
1199 |
return nr_free_zone_pages(gfp_zone(GFP_USER)); |
1da177e4c Linux-2.6.12-rc2 |
1200 1201 1202 1203 1204 1205 1206 |
} /* * Amount of free RAM allocatable within all zones */ unsigned int nr_free_pagecache_pages(void) { |
af4ca457e [PATCH] gfp_t: in... |
1207 |
return nr_free_zone_pages(gfp_zone(GFP_HIGHUSER)); |
1da177e4c Linux-2.6.12-rc2 |
1208 |
} |
08e0f6a97 [PATCH] Add NUMA_... |
1209 1210 |
static inline void show_node(struct zone *zone) |
1da177e4c Linux-2.6.12-rc2 |
1211 |
{ |
08e0f6a97 [PATCH] Add NUMA_... |
1212 1213 |
if (NUMA_BUILD) printk("Node %ld ", zone_to_nid(zone)); |
1da177e4c Linux-2.6.12-rc2 |
1214 |
} |
1da177e4c Linux-2.6.12-rc2 |
1215 |
|
1da177e4c Linux-2.6.12-rc2 |
1216 1217 1218 1219 1220 1221 |
void si_meminfo(struct sysinfo *val) { val->totalram = totalram_pages; val->sharedram = 0; val->freeram = nr_free_pages(); val->bufferram = nr_blockdev_pages(); |
1da177e4c Linux-2.6.12-rc2 |
1222 1223 |
val->totalhigh = totalhigh_pages; val->freehigh = nr_free_highpages(); |
1da177e4c Linux-2.6.12-rc2 |
1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 |
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; val->freeram = nr_free_pages_pgdat(pgdat); |
98d2b0ebd [PATCH] reduce MA... |
1236 |
#ifdef CONFIG_HIGHMEM |
1da177e4c Linux-2.6.12-rc2 |
1237 1238 |
val->totalhigh = pgdat->node_zones[ZONE_HIGHMEM].present_pages; val->freehigh = pgdat->node_zones[ZONE_HIGHMEM].free_pages; |
98d2b0ebd [PATCH] reduce MA... |
1239 1240 1241 1242 |
#else val->totalhigh = 0; val->freehigh = 0; #endif |
1da177e4c Linux-2.6.12-rc2 |
1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 |
val->mem_unit = PAGE_SIZE; } #endif #define K(x) ((x) << (PAGE_SHIFT-10)) /* * 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. */ void show_free_areas(void) { |
c72419138 [PATCH] Condense ... |
1256 |
int cpu; |
1da177e4c Linux-2.6.12-rc2 |
1257 1258 1259 1260 1261 1262 |
unsigned long active; unsigned long inactive; unsigned long free; struct zone *zone; for_each_zone(zone) { |
c72419138 [PATCH] Condense ... |
1263 |
if (!populated_zone(zone)) |
1da177e4c Linux-2.6.12-rc2 |
1264 |
continue; |
c72419138 [PATCH] Condense ... |
1265 1266 1267 1268 |
show_node(zone); printk("%s per-cpu: ", zone->name); |
1da177e4c Linux-2.6.12-rc2 |
1269 |
|
6b482c677 [PATCH] Don't pri... |
1270 |
for_each_online_cpu(cpu) { |
1da177e4c Linux-2.6.12-rc2 |
1271 |
struct per_cpu_pageset *pageset; |
e7c8d5c99 [PATCH] node loca... |
1272 |
pageset = zone_pcp(zone, cpu); |
1da177e4c Linux-2.6.12-rc2 |
1273 |
|
c72419138 [PATCH] Condense ... |
1274 1275 1276 1277 1278 1279 1280 |
printk("CPU %4d: Hot: hi:%5d, btch:%4d usd:%4d " "Cold: hi:%5d, btch:%4d usd:%4d ", cpu, pageset->pcp[0].high, pageset->pcp[0].batch, pageset->pcp[0].count, pageset->pcp[1].high, pageset->pcp[1].batch, pageset->pcp[1].count); |
1da177e4c Linux-2.6.12-rc2 |
1281 1282 |
} } |
1da177e4c Linux-2.6.12-rc2 |
1283 |
get_zone_counts(&active, &inactive, &free); |
1da177e4c Linux-2.6.12-rc2 |
1284 1285 1286 1287 1288 |
printk("Active:%lu inactive:%lu dirty:%lu writeback:%lu " "unstable:%lu free:%u slab:%lu mapped:%lu pagetables:%lu ", active, inactive, |
b1e7a8fd8 [PATCH] zoned vm ... |
1289 |
global_page_state(NR_FILE_DIRTY), |
ce866b34a [PATCH] zoned vm ... |
1290 |
global_page_state(NR_WRITEBACK), |
fd39fc856 [PATCH] zoned vm ... |
1291 |
global_page_state(NR_UNSTABLE_NFS), |
1da177e4c Linux-2.6.12-rc2 |
1292 |
nr_free_pages(), |
972d1a7b1 [PATCH] ZVC: Supp... |
1293 1294 |
global_page_state(NR_SLAB_RECLAIMABLE) + global_page_state(NR_SLAB_UNRECLAIMABLE), |
65ba55f50 [PATCH] zoned vm ... |
1295 |
global_page_state(NR_FILE_MAPPED), |
df849a152 [PATCH] zoned vm ... |
1296 |
global_page_state(NR_PAGETABLE)); |
1da177e4c Linux-2.6.12-rc2 |
1297 1298 1299 |
for_each_zone(zone) { int i; |
c72419138 [PATCH] Condense ... |
1300 1301 |
if (!populated_zone(zone)) continue; |
1da177e4c Linux-2.6.12-rc2 |
1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 |
show_node(zone); printk("%s" " free:%lukB" " min:%lukB" " low:%lukB" " high:%lukB" " active:%lukB" " inactive:%lukB" " present:%lukB" " pages_scanned:%lu" " all_unreclaimable? %s" " ", zone->name, K(zone->free_pages), K(zone->pages_min), K(zone->pages_low), K(zone->pages_high), K(zone->nr_active), K(zone->nr_inactive), K(zone->present_pages), zone->pages_scanned, (zone->all_unreclaimable ? "yes" : "no") ); printk("lowmem_reserve[]:"); for (i = 0; i < MAX_NR_ZONES; i++) printk(" %lu", zone->lowmem_reserve[i]); printk(" "); } for_each_zone(zone) { |
8f9de51a4 [PATCH] printk() ... |
1334 |
unsigned long nr[MAX_ORDER], flags, order, total = 0; |
1da177e4c Linux-2.6.12-rc2 |
1335 |
|
c72419138 [PATCH] Condense ... |
1336 1337 |
if (!populated_zone(zone)) continue; |
1da177e4c Linux-2.6.12-rc2 |
1338 1339 |
show_node(zone); printk("%s: ", zone->name); |
1da177e4c Linux-2.6.12-rc2 |
1340 1341 1342 |
spin_lock_irqsave(&zone->lock, flags); for (order = 0; order < MAX_ORDER; order++) { |
8f9de51a4 [PATCH] printk() ... |
1343 1344 |
nr[order] = zone->free_area[order].nr_free; total += nr[order] << order; |
1da177e4c Linux-2.6.12-rc2 |
1345 1346 |
} spin_unlock_irqrestore(&zone->lock, flags); |
8f9de51a4 [PATCH] printk() ... |
1347 1348 |
for (order = 0; order < MAX_ORDER; order++) printk("%lu*%lukB ", nr[order], K(1UL) << order); |
1da177e4c Linux-2.6.12-rc2 |
1349 1350 1351 1352 1353 1354 1355 1356 1357 |
printk("= %lukB ", K(total)); } show_swap_cache_info(); } /* * Builds allocation fallback zone lists. |
1a93205bd [PATCH] mm: simpl... |
1358 1359 |
* * Add all populated zones of a node to the zonelist. |
1da177e4c Linux-2.6.12-rc2 |
1360 |
*/ |
86356ab14 [PATCH] wait_tabl... |
1361 |
static int __meminit build_zonelists_node(pg_data_t *pgdat, |
2f6726e54 [PATCH] Apply typ... |
1362 |
struct zonelist *zonelist, int nr_zones, enum zone_type zone_type) |
1da177e4c Linux-2.6.12-rc2 |
1363 |
{ |
1a93205bd [PATCH] mm: simpl... |
1364 |
struct zone *zone; |
98d2b0ebd [PATCH] reduce MA... |
1365 |
BUG_ON(zone_type >= MAX_NR_ZONES); |
2f6726e54 [PATCH] Apply typ... |
1366 |
zone_type++; |
02a68a5eb [PATCH] Fix zone ... |
1367 1368 |
do { |
2f6726e54 [PATCH] Apply typ... |
1369 |
zone_type--; |
070f80326 [PATCH] build_zon... |
1370 |
zone = pgdat->node_zones + zone_type; |
1a93205bd [PATCH] mm: simpl... |
1371 |
if (populated_zone(zone)) { |
070f80326 [PATCH] build_zon... |
1372 1373 |
zonelist->zones[nr_zones++] = zone; check_highest_zone(zone_type); |
1da177e4c Linux-2.6.12-rc2 |
1374 |
} |
02a68a5eb [PATCH] Fix zone ... |
1375 |
|
2f6726e54 [PATCH] Apply typ... |
1376 |
} while (zone_type); |
070f80326 [PATCH] build_zon... |
1377 |
return nr_zones; |
1da177e4c Linux-2.6.12-rc2 |
1378 1379 1380 1381 |
} #ifdef CONFIG_NUMA #define MAX_NODE_LOAD (num_online_nodes()) |
86356ab14 [PATCH] wait_tabl... |
1382 |
static int __meminitdata node_load[MAX_NUMNODES]; |
1da177e4c Linux-2.6.12-rc2 |
1383 |
/** |
4dc3b16ba [PATCH] DocBook: ... |
1384 |
* find_next_best_node - find the next node that should appear in a given node's fallback list |
1da177e4c Linux-2.6.12-rc2 |
1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 |
* @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. */ |
86356ab14 [PATCH] wait_tabl... |
1397 |
static int __meminit find_next_best_node(int node, nodemask_t *used_node_mask) |
1da177e4c Linux-2.6.12-rc2 |
1398 |
{ |
4cf808eb4 [PATCH] Handle ho... |
1399 |
int n, val; |
1da177e4c Linux-2.6.12-rc2 |
1400 1401 |
int min_val = INT_MAX; int best_node = -1; |
4cf808eb4 [PATCH] Handle ho... |
1402 1403 1404 1405 1406 |
/* 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 |
1407 |
|
4cf808eb4 [PATCH] Handle ho... |
1408 1409 |
for_each_online_node(n) { cpumask_t tmp; |
1da177e4c Linux-2.6.12-rc2 |
1410 1411 1412 1413 |
/* Don't want a node to appear more than once */ if (node_isset(n, *used_node_mask)) continue; |
1da177e4c Linux-2.6.12-rc2 |
1414 1415 |
/* Use the distance array to find the distance */ val = node_distance(node, n); |
4cf808eb4 [PATCH] Handle ho... |
1416 1417 |
/* Penalize nodes under us ("prefer the next node") */ val += (n < node); |
1da177e4c Linux-2.6.12-rc2 |
1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 |
/* Give preference to headless and unused nodes */ tmp = node_to_cpumask(n); if (!cpus_empty(tmp)) 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; } |
86356ab14 [PATCH] wait_tabl... |
1438 |
static void __meminit build_zonelists(pg_data_t *pgdat) |
1da177e4c Linux-2.6.12-rc2 |
1439 |
{ |
19655d348 [PATCH] linearly ... |
1440 1441 |
int j, node, local_node; enum zone_type i; |
1da177e4c Linux-2.6.12-rc2 |
1442 1443 1444 1445 1446 |
int prev_node, load; struct zonelist *zonelist; nodemask_t used_mask; /* initialize zonelists */ |
19655d348 [PATCH] linearly ... |
1447 |
for (i = 0; i < MAX_NR_ZONES; i++) { |
1da177e4c Linux-2.6.12-rc2 |
1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 |
zonelist = pgdat->node_zonelists + i; zonelist->zones[0] = NULL; } /* NUMA-aware ordering of nodes */ local_node = pgdat->node_id; load = num_online_nodes(); prev_node = local_node; nodes_clear(used_mask); while ((node = find_next_best_node(local_node, &used_mask)) >= 0) { |
9eeff2395 [PATCH] Zone recl... |
1458 1459 1460 1461 1462 1463 1464 1465 |
int distance = node_distance(local_node, node); /* * If another node is sufficiently far away then it is better * to reclaim pages in a zone before going off node. */ if (distance > RECLAIM_DISTANCE) zone_reclaim_mode = 1; |
1da177e4c Linux-2.6.12-rc2 |
1466 1467 1468 1469 1470 |
/* * 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. */ |
9eeff2395 [PATCH] Zone recl... |
1471 1472 |
if (distance != node_distance(local_node, prev_node)) |
1da177e4c Linux-2.6.12-rc2 |
1473 1474 1475 |
node_load[node] += load; prev_node = node; load--; |
19655d348 [PATCH] linearly ... |
1476 |
for (i = 0; i < MAX_NR_ZONES; i++) { |
1da177e4c Linux-2.6.12-rc2 |
1477 1478 |
zonelist = pgdat->node_zonelists + i; for (j = 0; zonelist->zones[j] != NULL; j++); |
19655d348 [PATCH] linearly ... |
1479 |
j = build_zonelists_node(NODE_DATA(node), zonelist, j, i); |
1da177e4c Linux-2.6.12-rc2 |
1480 1481 1482 1483 1484 1485 |
zonelist->zones[j] = NULL; } } } #else /* CONFIG_NUMA */ |
86356ab14 [PATCH] wait_tabl... |
1486 |
static void __meminit build_zonelists(pg_data_t *pgdat) |
1da177e4c Linux-2.6.12-rc2 |
1487 |
{ |
19655d348 [PATCH] linearly ... |
1488 1489 |
int node, local_node; enum zone_type i,j; |
1da177e4c Linux-2.6.12-rc2 |
1490 1491 |
local_node = pgdat->node_id; |
19655d348 [PATCH] linearly ... |
1492 |
for (i = 0; i < MAX_NR_ZONES; i++) { |
1da177e4c Linux-2.6.12-rc2 |
1493 1494 1495 |
struct zonelist *zonelist; zonelist = pgdat->node_zonelists + i; |
19655d348 [PATCH] linearly ... |
1496 |
j = build_zonelists_node(pgdat, zonelist, 0, i); |
1da177e4c Linux-2.6.12-rc2 |
1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 |
/* * 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; |
19655d348 [PATCH] linearly ... |
1508 |
j = build_zonelists_node(NODE_DATA(node), zonelist, j, i); |
1da177e4c Linux-2.6.12-rc2 |
1509 1510 1511 1512 |
} for (node = 0; node < local_node; node++) { if (!node_online(node)) continue; |
19655d348 [PATCH] linearly ... |
1513 |
j = build_zonelists_node(NODE_DATA(node), zonelist, j, i); |
1da177e4c Linux-2.6.12-rc2 |
1514 1515 1516 1517 1518 1519 1520 |
} zonelist->zones[j] = NULL; } } #endif /* CONFIG_NUMA */ |
6811378e7 [PATCH] wait_tabl... |
1521 1522 |
/* return values int ....just for stop_machine_run() */ static int __meminit __build_all_zonelists(void *dummy) |
1da177e4c Linux-2.6.12-rc2 |
1523 |
{ |
6811378e7 [PATCH] wait_tabl... |
1524 1525 1526 1527 1528 1529 1530 1531 1532 |
int nid; for_each_online_node(nid) build_zonelists(NODE_DATA(nid)); return 0; } void __meminit build_all_zonelists(void) { if (system_state == SYSTEM_BOOTING) { |
423b41d77 [PATCH] mm/page_a... |
1533 |
__build_all_zonelists(NULL); |
6811378e7 [PATCH] wait_tabl... |
1534 1535 1536 1537 1538 1539 1540 |
cpuset_init_current_mems_allowed(); } else { /* we have to stop all cpus to guaranntee there is no user of zonelist */ stop_machine_run(__build_all_zonelists, NULL, NR_CPUS); /* cpuset refresh routine should be here */ } |
bd1e22b8e [PATCH] initialis... |
1541 1542 1543 1544 |
vm_total_pages = nr_free_pagecache_pages(); printk("Built %i zonelists. Total pages: %ld ", num_online_nodes(), vm_total_pages); |
1da177e4c Linux-2.6.12-rc2 |
1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 |
} /* * 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... |
1559 |
#ifndef CONFIG_MEMORY_HOTPLUG |
02b694dea [PATCH] wait_tabl... |
1560 |
static inline unsigned long wait_table_hash_nr_entries(unsigned long pages) |
1da177e4c Linux-2.6.12-rc2 |
1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 |
{ 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... |
1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 |
#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 |
1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 |
/* * 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)) |
1da177e4c Linux-2.6.12-rc2 |
1613 1614 1615 1616 1617 |
/* * 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... |
1618 |
void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, |
1da177e4c Linux-2.6.12-rc2 |
1619 1620 |
unsigned long start_pfn) { |
1da177e4c Linux-2.6.12-rc2 |
1621 |
struct page *page; |
29751f699 [PATCH] sparsemem... |
1622 1623 |
unsigned long end_pfn = start_pfn + size; unsigned long pfn; |
1da177e4c Linux-2.6.12-rc2 |
1624 |
|
cbe8dd4af [PATCH] memmap_in... |
1625 |
for (pfn = start_pfn; pfn < end_pfn; pfn++) { |
d41dee369 [PATCH] sparsemem... |
1626 1627 1628 1629 |
if (!early_pfn_valid(pfn)) continue; page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); |
7835e98b2 [PATCH] remove se... |
1630 |
init_page_count(page); |
1da177e4c Linux-2.6.12-rc2 |
1631 1632 1633 1634 1635 1636 |
reset_page_mapcount(page); SetPageReserved(page); 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_... |
1637 |
set_page_address(page, __va(pfn << PAGE_SHIFT)); |
1da177e4c Linux-2.6.12-rc2 |
1638 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 |
} } void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone, unsigned long size) { int order; for (order = 0; order < MAX_ORDER ; order++) { INIT_LIST_HEAD(&zone->free_area[order].free_list); zone->free_area[order].nr_free = 0; } } |
d41dee369 [PATCH] sparsemem... |
1651 |
#define ZONETABLE_INDEX(x, zone_nr) ((x << ZONES_SHIFT) | zone_nr) |
2f1b62486 [PATCH] reduce MA... |
1652 1653 |
void zonetable_add(struct zone *zone, int nid, enum zone_type zid, unsigned long pfn, unsigned long size) |
d41dee369 [PATCH] sparsemem... |
1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 |
{ unsigned long snum = pfn_to_section_nr(pfn); unsigned long end = pfn_to_section_nr(pfn + size); if (FLAGS_HAS_NODE) zone_table[ZONETABLE_INDEX(nid, zid)] = zone; else for (; snum <= end; snum++) zone_table[ZONETABLE_INDEX(snum, zid)] = zone; } |
1da177e4c Linux-2.6.12-rc2 |
1664 1665 1666 1667 |
#ifndef __HAVE_ARCH_MEMMAP_INIT #define memmap_init(size, nid, zone, start_pfn) \ memmap_init_zone((size), (nid), (zone), (start_pfn)) #endif |
6292d9aaf [PATCH] __cpuinit... |
1668 |
static int __cpuinit zone_batchsize(struct zone *zone) |
e7c8d5c99 [PATCH] node loca... |
1669 1670 1671 1672 1673 |
{ int batch; /* * The per-cpu-pages pools are set to around 1000th of the |
ba56e91c9 [PATCH] mm: page_... |
1674 |
* size of the zone. But no more than 1/2 of a meg. |
e7c8d5c99 [PATCH] node loca... |
1675 1676 1677 1678 |
* * OK, so we don't know how big the cache is. So guess. */ batch = zone->present_pages / 1024; |
ba56e91c9 [PATCH] mm: page_... |
1679 1680 |
if (batch * PAGE_SIZE > 512 * 1024) batch = (512 * 1024) / PAGE_SIZE; |
e7c8d5c99 [PATCH] node loca... |
1681 1682 1683 1684 1685 |
batch /= 4; /* We effectively *= 4 below */ if (batch < 1) batch = 1; /* |
0ceaacc97 [PATCH] Fix up pe... |
1686 1687 1688 |
* 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... |
1689 |
* |
0ceaacc97 [PATCH] Fix up pe... |
1690 1691 1692 1693 |
* 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... |
1694 |
*/ |
0ceaacc97 [PATCH] Fix up pe... |
1695 |
batch = (1 << (fls(batch + batch/2)-1)) - 1; |
ba56e91c9 [PATCH] mm: page_... |
1696 |
|
e7c8d5c99 [PATCH] node loca... |
1697 1698 |
return batch; } |
2caaad41e [PATCH] Reduce si... |
1699 1700 1701 |
inline void setup_pageset(struct per_cpu_pageset *p, unsigned long batch) { struct per_cpu_pages *pcp; |
1c6fe9465 [PATCH] NUMA: bro... |
1702 |
memset(p, 0, sizeof(*p)); |
2caaad41e [PATCH] Reduce si... |
1703 1704 |
pcp = &p->pcp[0]; /* hot */ pcp->count = 0; |
2caaad41e [PATCH] Reduce si... |
1705 1706 1707 1708 1709 1710 |
pcp->high = 6 * batch; pcp->batch = max(1UL, 1 * batch); INIT_LIST_HEAD(&pcp->list); pcp = &p->pcp[1]; /* cold*/ pcp->count = 0; |
2caaad41e [PATCH] Reduce si... |
1711 |
pcp->high = 2 * batch; |
e46a5e28c [PATCH] mm: set p... |
1712 |
pcp->batch = max(1UL, batch/2); |
2caaad41e [PATCH] Reduce si... |
1713 1714 |
INIT_LIST_HEAD(&pcp->list); } |
8ad4b1fb8 [PATCH] Make high... |
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 |
/* * 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; pcp = &p->pcp[0]; /* hot list */ pcp->high = high; pcp->batch = max(1UL, high/4); if ((high/4) > (PAGE_SHIFT * 8)) pcp->batch = PAGE_SHIFT * 8; } |
e7c8d5c99 [PATCH] node loca... |
1731 1732 |
#ifdef CONFIG_NUMA /* |
2caaad41e [PATCH] Reduce si... |
1733 1734 1735 1736 1737 1738 1739 |
* 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. * * Some NUMA counter updates may also be caught by the boot pagesets. |
b7c84c6ad [PATCH] boot_page... |
1740 1741 1742 1743 1744 1745 1746 1747 |
* * 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. |
2caaad41e [PATCH] Reduce si... |
1748 |
*/ |
88a2a4ac6 [PATCH] percpu da... |
1749 |
static struct per_cpu_pageset boot_pageset[NR_CPUS]; |
2caaad41e [PATCH] Reduce si... |
1750 1751 1752 |
/* * Dynamically allocate memory for the |
e7c8d5c99 [PATCH] node loca... |
1753 1754 |
* per cpu pageset array in struct zone. */ |
6292d9aaf [PATCH] __cpuinit... |
1755 |
static int __cpuinit process_zones(int cpu) |
e7c8d5c99 [PATCH] node loca... |
1756 1757 |
{ struct zone *zone, *dzone; |
e7c8d5c99 [PATCH] node loca... |
1758 1759 |
for_each_zone(zone) { |
e7c8d5c99 [PATCH] node loca... |
1760 |
|
66a550308 [PATCH] Do not al... |
1761 1762 |
if (!populated_zone(zone)) continue; |
23316bc86 [PATCH] mm: clean... |
1763 |
zone_pcp(zone, cpu) = kmalloc_node(sizeof(struct per_cpu_pageset), |
e7c8d5c99 [PATCH] node loca... |
1764 |
GFP_KERNEL, cpu_to_node(cpu)); |
23316bc86 [PATCH] mm: clean... |
1765 |
if (!zone_pcp(zone, cpu)) |
e7c8d5c99 [PATCH] node loca... |
1766 |
goto bad; |
e7c8d5c99 [PATCH] node loca... |
1767 |
|
23316bc86 [PATCH] mm: clean... |
1768 |
setup_pageset(zone_pcp(zone, cpu), zone_batchsize(zone)); |
8ad4b1fb8 [PATCH] Make high... |
1769 1770 1771 1772 |
if (percpu_pagelist_fraction) setup_pagelist_highmark(zone_pcp(zone, cpu), (zone->present_pages / percpu_pagelist_fraction)); |
e7c8d5c99 [PATCH] node loca... |
1773 1774 1775 1776 1777 1778 1779 |
} return 0; bad: for_each_zone(dzone) { if (dzone == zone) break; |
23316bc86 [PATCH] mm: clean... |
1780 1781 |
kfree(zone_pcp(dzone, cpu)); zone_pcp(dzone, cpu) = NULL; |
e7c8d5c99 [PATCH] node loca... |
1782 1783 1784 1785 1786 1787 |
} return -ENOMEM; } static inline void free_zone_pagesets(int cpu) { |
e7c8d5c99 [PATCH] node loca... |
1788 1789 1790 1791 |
struct zone *zone; for_each_zone(zone) { struct per_cpu_pageset *pset = zone_pcp(zone, cpu); |
f3ef9ead3 [PATCH] do not fr... |
1792 1793 1794 |
/* Free per_cpu_pageset if it is slab allocated */ if (pset != &boot_pageset[cpu]) kfree(pset); |
e7c8d5c99 [PATCH] node loca... |
1795 |
zone_pcp(zone, cpu) = NULL; |
e7c8d5c99 [PATCH] node loca... |
1796 |
} |
e7c8d5c99 [PATCH] node loca... |
1797 |
} |
9c7b216d2 [PATCH] cpu hotpl... |
1798 |
static int __cpuinit pageset_cpuup_callback(struct notifier_block *nfb, |
e7c8d5c99 [PATCH] node loca... |
1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 |
unsigned long action, void *hcpu) { int cpu = (long)hcpu; int ret = NOTIFY_OK; switch (action) { case CPU_UP_PREPARE: if (process_zones(cpu)) ret = NOTIFY_BAD; break; |
b0d416932 [PATCH] x86_64: W... |
1810 |
case CPU_UP_CANCELED: |
e7c8d5c99 [PATCH] node loca... |
1811 1812 1813 |
case CPU_DEAD: free_zone_pagesets(cpu); break; |
e7c8d5c99 [PATCH] node loca... |
1814 1815 1816 1817 1818 |
default: break; } return ret; } |
74b85f379 [PATCH] cpu hotpl... |
1819 |
static struct notifier_block __cpuinitdata pageset_notifier = |
e7c8d5c99 [PATCH] node loca... |
1820 |
{ &pageset_cpuup_callback, NULL, 0 }; |
78d9955bb [PATCH] missing p... |
1821 |
void __init setup_per_cpu_pageset(void) |
e7c8d5c99 [PATCH] node loca... |
1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 |
{ int err; /* Initialize per_cpu_pageset for cpu 0. * A cpuup callback will do this for every cpu * as it comes online */ err = process_zones(smp_processor_id()); BUG_ON(err); register_cpu_notifier(&pageset_notifier); } #endif |
c09b42404 [PATCH] x86_64: a... |
1835 |
static __meminit |
cca448fe9 [PATCH] wait_tabl... |
1836 |
int zone_wait_table_init(struct zone *zone, unsigned long zone_size_pages) |
ed8ece2ec [PATCH] memory ho... |
1837 1838 1839 |
{ int i; struct pglist_data *pgdat = zone->zone_pgdat; |
cca448fe9 [PATCH] wait_tabl... |
1840 |
size_t alloc_size; |
ed8ece2ec [PATCH] memory ho... |
1841 1842 1843 1844 1845 |
/* * The per-page waitqueue mechanism uses hashed waitqueues * per zone. */ |
02b694dea [PATCH] wait_tabl... |
1846 1847 1848 1849 |
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... |
1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 |
alloc_size = zone->wait_table_hash_nr_entries * sizeof(wait_queue_head_t); if (system_state == SYSTEM_BOOTING) { zone->wait_table = (wait_queue_head_t *) alloc_bootmem_node(pgdat, alloc_size); } 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. */ zone->wait_table = (wait_queue_head_t *)vmalloc(alloc_size); } if (!zone->wait_table) return -ENOMEM; |
ed8ece2ec [PATCH] memory ho... |
1871 |
|
02b694dea [PATCH] wait_tabl... |
1872 |
for(i = 0; i < zone->wait_table_hash_nr_entries; ++i) |
ed8ece2ec [PATCH] memory ho... |
1873 |
init_waitqueue_head(zone->wait_table + i); |
cca448fe9 [PATCH] wait_tabl... |
1874 1875 |
return 0; |
ed8ece2ec [PATCH] memory ho... |
1876 |
} |
c09b42404 [PATCH] x86_64: a... |
1877 |
static __meminit void zone_pcp_init(struct zone *zone) |
ed8ece2ec [PATCH] memory ho... |
1878 1879 1880 1881 1882 1883 1884 |
{ int cpu; unsigned long batch = zone_batchsize(zone); for (cpu = 0; cpu < NR_CPUS; cpu++) { #ifdef CONFIG_NUMA /* Early boot. Slab allocator not functional yet */ |
23316bc86 [PATCH] mm: clean... |
1885 |
zone_pcp(zone, cpu) = &boot_pageset[cpu]; |
ed8ece2ec [PATCH] memory ho... |
1886 1887 1888 1889 1890 |
setup_pageset(&boot_pageset[cpu],0); #else setup_pageset(zone_pcp(zone,cpu), batch); #endif } |
f5335c0f1 [PATCH] quieten z... |
1891 1892 1893 1894 |
if (zone->present_pages) printk(KERN_DEBUG " %s zone: %lu pages, LIFO batch:%lu ", zone->name, zone->present_pages, batch); |
ed8ece2ec [PATCH] memory ho... |
1895 |
} |
718127cc3 [PATCH] wait_tabl... |
1896 1897 1898 |
__meminit int init_currently_empty_zone(struct zone *zone, unsigned long zone_start_pfn, unsigned long size) |
ed8ece2ec [PATCH] memory ho... |
1899 1900 |
{ struct pglist_data *pgdat = zone->zone_pgdat; |
cca448fe9 [PATCH] wait_tabl... |
1901 1902 1903 1904 |
int ret; ret = zone_wait_table_init(zone, size); if (ret) return ret; |
ed8ece2ec [PATCH] memory ho... |
1905 |
pgdat->nr_zones = zone_idx(zone) + 1; |
ed8ece2ec [PATCH] memory ho... |
1906 1907 1908 1909 1910 |
zone->zone_start_pfn = zone_start_pfn; memmap_init(size, pgdat->node_id, zone_idx(zone), zone_start_pfn); zone_init_free_lists(pgdat, zone, zone->spanned_pages); |
718127cc3 [PATCH] wait_tabl... |
1911 1912 |
return 0; |
ed8ece2ec [PATCH] memory ho... |
1913 |
} |
c713216de [PATCH] Introduce... |
1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 |
#ifdef CONFIG_ARCH_POPULATES_NODE_MAP /* * Basic iterator support. Return the first range of PFNs for a node * Note: nid == MAX_NUMNODES returns first region regardless of node */ static int __init first_active_region_index_in_nid(int nid) { int i; for (i = 0; i < nr_nodemap_entries; i++) if (nid == MAX_NUMNODES || early_node_map[i].nid == nid) return i; return -1; } /* * Basic iterator support. Return the next active range of PFNs for a node * Note: nid == MAX_NUMNODES returns next region regardles of node */ static int __init next_active_region_index_in_nid(int index, int nid) { for (index = index + 1; index < nr_nodemap_entries; index++) if (nid == MAX_NUMNODES || early_node_map[index].nid == nid) return index; return -1; } #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 */ int __init early_pfn_to_nid(unsigned long pfn) { int i; for (i = 0; i < nr_nodemap_entries; i++) { unsigned long start_pfn = early_node_map[i].start_pfn; unsigned long end_pfn = early_node_map[i].end_pfn; if (start_pfn <= pfn && pfn < end_pfn) return early_node_map[i].nid; } return 0; } #endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ /* Basic iterator support to walk early_node_map[] */ #define for_each_active_range_index_in_nid(i, nid) \ for (i = first_active_region_index_in_nid(nid); i != -1; \ i = next_active_region_index_in_nid(i, nid)) /** * free_bootmem_with_active_regions - Call free_bootmem_node for each active range |
88ca3b94e [PATCH] page_allo... |
1973 1974 |
* @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... |
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 |
* * 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. */ void __init free_bootmem_with_active_regions(int nid, unsigned long max_low_pfn) { int i; for_each_active_range_index_in_nid(i, nid) { unsigned long size_pages = 0; unsigned long end_pfn = early_node_map[i].end_pfn; if (early_node_map[i].start_pfn >= max_low_pfn) continue; if (end_pfn > max_low_pfn) end_pfn = max_low_pfn; size_pages = end_pfn - early_node_map[i].start_pfn; free_bootmem_node(NODE_DATA(early_node_map[i].nid), PFN_PHYS(early_node_map[i].start_pfn), size_pages << PAGE_SHIFT); } } /** * sparse_memory_present_with_active_regions - Call memory_present for each active range |
88ca3b94e [PATCH] page_allo... |
2004 |
* @nid: The node to call memory_present for. If MAX_NUMNODES, all nodes will be used. |
c713216de [PATCH] Introduce... |
2005 2006 2007 |
* * If an architecture guarantees that all ranges registered with * add_active_ranges() contain no holes and may be freed, this |
88ca3b94e [PATCH] page_allo... |
2008 |
* function may be used instead of calling memory_present() manually. |
c713216de [PATCH] Introduce... |
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 |
*/ void __init sparse_memory_present_with_active_regions(int nid) { int i; for_each_active_range_index_in_nid(i, nid) memory_present(early_node_map[i].nid, early_node_map[i].start_pfn, early_node_map[i].end_pfn); } /** |
fb01439c5 [PATCH] Allow an ... |
2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 |
* push_node_boundaries - Push node boundaries to at least the requested boundary * @nid: The nid of the node to push the boundary for * @start_pfn: The start pfn of the node * @end_pfn: The end pfn of the node * * In reserve-based hot-add, mem_map is allocated that is unused until hotadd * time. Specifically, on x86_64, SRAT will report ranges that can potentially * be hotplugged even though no physical memory exists. This function allows * an arch to push out the node boundaries so mem_map is allocated that can * be used later. */ #ifdef CONFIG_MEMORY_HOTPLUG_RESERVE void __init push_node_boundaries(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn) { printk(KERN_DEBUG "Entering push_node_boundaries(%u, %lu, %lu) ", nid, start_pfn, end_pfn); /* Initialise the boundary for this node if necessary */ if (node_boundary_end_pfn[nid] == 0) node_boundary_start_pfn[nid] = -1UL; /* Update the boundaries */ if (node_boundary_start_pfn[nid] > start_pfn) node_boundary_start_pfn[nid] = start_pfn; if (node_boundary_end_pfn[nid] < end_pfn) node_boundary_end_pfn[nid] = end_pfn; } /* If necessary, push the node boundary out for reserve hotadd */ static void __init account_node_boundary(unsigned int nid, unsigned long *start_pfn, unsigned long *end_pfn) { printk(KERN_DEBUG "Entering account_node_boundary(%u, %lu, %lu) ", nid, *start_pfn, *end_pfn); /* Return if boundary information has not been provided */ if (node_boundary_end_pfn[nid] == 0) return; /* Check the boundaries and update if necessary */ if (node_boundary_start_pfn[nid] < *start_pfn) *start_pfn = node_boundary_start_pfn[nid]; if (node_boundary_end_pfn[nid] > *end_pfn) *end_pfn = node_boundary_end_pfn[nid]; } #else void __init push_node_boundaries(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn) {} static void __init account_node_boundary(unsigned int nid, unsigned long *start_pfn, unsigned long *end_pfn) {} #endif /** |
c713216de [PATCH] Introduce... |
2079 |
* get_pfn_range_for_nid - Return the start and end page frames for a node |
88ca3b94e [PATCH] page_allo... |
2080 2081 2082 |
* @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... |
2083 2084 2085 2086 |
* * 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... |
2087 |
* PFNs will be 0. |
c713216de [PATCH] Introduce... |
2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 |
*/ void __init get_pfn_range_for_nid(unsigned int nid, unsigned long *start_pfn, unsigned long *end_pfn) { int i; *start_pfn = -1UL; *end_pfn = 0; for_each_active_range_index_in_nid(i, nid) { *start_pfn = min(*start_pfn, early_node_map[i].start_pfn); *end_pfn = max(*end_pfn, early_node_map[i].end_pfn); } if (*start_pfn == -1UL) { printk(KERN_WARNING "Node %u active with no memory ", nid); *start_pfn = 0; } |
fb01439c5 [PATCH] Allow an ... |
2106 2107 2108 |
/* Push the node boundaries out if requested */ account_node_boundary(nid, start_pfn, end_pfn); |
c713216de [PATCH] Introduce... |
2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 |
} /* * 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() */ unsigned long __init zone_spanned_pages_in_node(int nid, 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]; /* 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... |
2141 |
* then all holes in the requested range will be accounted for. |
c713216de [PATCH] Introduce... |
2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 |
*/ unsigned long __init __absent_pages_in_range(int nid, unsigned long range_start_pfn, unsigned long range_end_pfn) { int i = 0; unsigned long prev_end_pfn = 0, hole_pages = 0; unsigned long start_pfn; /* Find the end_pfn of the first active range of pfns in the node */ i = first_active_region_index_in_nid(nid); if (i == -1) return 0; |
9c7cd6877 [PATCH] Account f... |
2155 2156 2157 |
/* Account for ranges before physical memory on this node */ if (early_node_map[i].start_pfn > range_start_pfn) hole_pages = early_node_map[i].start_pfn - range_start_pfn; |
c713216de [PATCH] Introduce... |
2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 |
prev_end_pfn = early_node_map[i].start_pfn; /* Find all holes for the zone within the node */ for (; i != -1; i = next_active_region_index_in_nid(i, nid)) { /* No need to continue if prev_end_pfn is outside the zone */ if (prev_end_pfn >= range_end_pfn) break; /* Make sure the end of the zone is not within the hole */ start_pfn = min(early_node_map[i].start_pfn, range_end_pfn); prev_end_pfn = max(prev_end_pfn, range_start_pfn); /* Update the hole size cound and move on */ if (start_pfn > range_start_pfn) { BUG_ON(prev_end_pfn > start_pfn); hole_pages += start_pfn - prev_end_pfn; } prev_end_pfn = early_node_map[i].end_pfn; } |
9c7cd6877 [PATCH] Account f... |
2178 2179 2180 2181 |
/* Account for ranges past physical memory on this node */ if (range_end_pfn > prev_end_pfn) hole_pages = range_end_pfn - max(range_start_pfn, prev_end_pfn); |
c713216de [PATCH] Introduce... |
2182 2183 2184 2185 2186 2187 2188 2189 |
return hole_pages; } /** * 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... |
2190 |
* It returns the number of pages frames in memory holes within a range. |
c713216de [PATCH] Introduce... |
2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 |
*/ 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 */ unsigned long __init zone_absent_pages_in_node(int nid, unsigned long zone_type, unsigned long *ignored) { |
9c7cd6877 [PATCH] Account f... |
2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 |
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); zone_start_pfn = max(arch_zone_lowest_possible_pfn[zone_type], node_start_pfn); zone_end_pfn = min(arch_zone_highest_possible_pfn[zone_type], node_end_pfn); return __absent_pages_in_range(nid, zone_start_pfn, zone_end_pfn); |
c713216de [PATCH] Introduce... |
2213 |
} |
0e0b864e0 [PATCH] Account f... |
2214 |
|
c713216de [PATCH] Introduce... |
2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 |
#else static inline unsigned long zone_spanned_pages_in_node(int nid, unsigned long zone_type, unsigned long *zones_size) { return zones_size[zone_type]; } static inline unsigned long zone_absent_pages_in_node(int nid, unsigned long zone_type, unsigned long *zholes_size) { if (!zholes_size) return 0; return zholes_size[zone_type]; } |
0e0b864e0 [PATCH] Account f... |
2232 |
|
c713216de [PATCH] Introduce... |
2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 |
#endif static void __init calculate_node_totalpages(struct pglist_data *pgdat, 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); } |
1da177e4c Linux-2.6.12-rc2 |
2256 2257 2258 2259 2260 2261 |
/* * Set up the zone data structures: * - mark all pages reserved * - mark all memory queues empty * - clear the memory bitmaps */ |
86356ab14 [PATCH] wait_tabl... |
2262 |
static void __meminit free_area_init_core(struct pglist_data *pgdat, |
1da177e4c Linux-2.6.12-rc2 |
2263 2264 |
unsigned long *zones_size, unsigned long *zholes_size) { |
2f1b62486 [PATCH] reduce MA... |
2265 |
enum zone_type j; |
ed8ece2ec [PATCH] memory ho... |
2266 |
int nid = pgdat->node_id; |
1da177e4c Linux-2.6.12-rc2 |
2267 |
unsigned long zone_start_pfn = pgdat->node_start_pfn; |
718127cc3 [PATCH] wait_tabl... |
2268 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
2269 |
|
208d54e55 [PATCH] memory ho... |
2270 |
pgdat_resize_init(pgdat); |
1da177e4c Linux-2.6.12-rc2 |
2271 2272 2273 2274 2275 2276 |
pgdat->nr_zones = 0; init_waitqueue_head(&pgdat->kswapd_wait); pgdat->kswapd_max_order = 0; for (j = 0; j < MAX_NR_ZONES; j++) { struct zone *zone = pgdat->node_zones + j; |
0e0b864e0 [PATCH] Account f... |
2277 |
unsigned long size, realsize, memmap_pages; |
1da177e4c Linux-2.6.12-rc2 |
2278 |
|
c713216de [PATCH] Introduce... |
2279 2280 2281 |
size = zone_spanned_pages_in_node(nid, j, zones_size); realsize = size - zone_absent_pages_in_node(nid, j, zholes_size); |
1da177e4c Linux-2.6.12-rc2 |
2282 |
|
0e0b864e0 [PATCH] Account f... |
2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 |
/* * Adjust realsize so that it accounts for how much memory * is used by this zone for memmap. This affects the watermark * and per-cpu initialisations */ memmap_pages = (size * sizeof(struct page)) >> PAGE_SHIFT; if (realsize >= memmap_pages) { realsize -= memmap_pages; printk(KERN_DEBUG " %s zone: %lu pages used for memmap ", zone_names[j], memmap_pages); } else printk(KERN_WARNING " %s zone: %lu pages exceeds realsize %lu ", zone_names[j], memmap_pages, realsize); /* Account for reserved DMA pages */ if (j == ZONE_DMA && realsize > dma_reserve) { realsize -= dma_reserve; printk(KERN_DEBUG " DMA zone: %lu pages reserved ", dma_reserve); } |
98d2b0ebd [PATCH] reduce MA... |
2308 |
if (!is_highmem_idx(j)) |
1da177e4c Linux-2.6.12-rc2 |
2309 2310 2311 2312 2313 |
nr_kernel_pages += realsize; nr_all_pages += realsize; zone->spanned_pages = size; zone->present_pages = realsize; |
9614634fe [PATCH] ZVC/zone_... |
2314 |
#ifdef CONFIG_NUMA |
d5f541ed6 [PATCH] Add node ... |
2315 |
zone->node = nid; |
8417bba4b [PATCH] Replace m... |
2316 |
zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio) |
9614634fe [PATCH] ZVC/zone_... |
2317 |
/ 100; |
0ff38490c [PATCH] zone_recl... |
2318 |
zone->min_slab_pages = (realsize * sysctl_min_slab_ratio) / 100; |
9614634fe [PATCH] ZVC/zone_... |
2319 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
2320 2321 2322 |
zone->name = zone_names[j]; spin_lock_init(&zone->lock); spin_lock_init(&zone->lru_lock); |
bdc8cb984 [PATCH] memory ho... |
2323 |
zone_seqlock_init(zone); |
1da177e4c Linux-2.6.12-rc2 |
2324 2325 2326 2327 |
zone->zone_pgdat = pgdat; zone->free_pages = 0; zone->temp_priority = zone->prev_priority = DEF_PRIORITY; |
ed8ece2ec [PATCH] memory ho... |
2328 |
zone_pcp_init(zone); |
1da177e4c Linux-2.6.12-rc2 |
2329 2330 2331 2332 2333 2334 |
INIT_LIST_HEAD(&zone->active_list); INIT_LIST_HEAD(&zone->inactive_list); zone->nr_scan_active = 0; zone->nr_scan_inactive = 0; zone->nr_active = 0; zone->nr_inactive = 0; |
2244b95a7 [PATCH] zoned vm ... |
2335 |
zap_zone_vm_stats(zone); |
53e9a6159 [PATCH] VM: zone ... |
2336 |
atomic_set(&zone->reclaim_in_progress, 0); |
1da177e4c Linux-2.6.12-rc2 |
2337 2338 |
if (!size) continue; |
d41dee369 [PATCH] sparsemem... |
2339 |
zonetable_add(zone, nid, j, zone_start_pfn, size); |
718127cc3 [PATCH] wait_tabl... |
2340 2341 |
ret = init_currently_empty_zone(zone, zone_start_pfn, size); BUG_ON(ret); |
1da177e4c Linux-2.6.12-rc2 |
2342 |
zone_start_pfn += size; |
1da177e4c Linux-2.6.12-rc2 |
2343 2344 2345 2346 2347 |
} } static void __init alloc_node_mem_map(struct pglist_data *pgdat) { |
1da177e4c Linux-2.6.12-rc2 |
2348 2349 2350 |
/* Skip empty nodes */ if (!pgdat->node_spanned_pages) return; |
d41dee369 [PATCH] sparsemem... |
2351 |
#ifdef CONFIG_FLAT_NODE_MEM_MAP |
1da177e4c Linux-2.6.12-rc2 |
2352 2353 |
/* ia64 gets its own node_mem_map, before this, without bootmem */ if (!pgdat->node_mem_map) { |
e984bb43f [PATCH] Align the... |
2354 |
unsigned long size, start, end; |
d41dee369 [PATCH] sparsemem... |
2355 |
struct page *map; |
e984bb43f [PATCH] Align the... |
2356 2357 2358 2359 2360 2361 2362 2363 2364 |
/* * 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); end = pgdat->node_start_pfn + pgdat->node_spanned_pages; end = ALIGN(end, MAX_ORDER_NR_PAGES); size = (end - start) * sizeof(struct page); |
6f167ec72 [PATCH] sparsemem... |
2365 2366 2367 |
map = alloc_remap(pgdat->node_id, size); if (!map) map = alloc_bootmem_node(pgdat, size); |
e984bb43f [PATCH] Align the... |
2368 |
pgdat->node_mem_map = map + (pgdat->node_start_pfn - start); |
1da177e4c Linux-2.6.12-rc2 |
2369 |
} |
d41dee369 [PATCH] sparsemem... |
2370 |
#ifdef CONFIG_FLATMEM |
1da177e4c Linux-2.6.12-rc2 |
2371 2372 2373 |
/* * With no DISCONTIG, the global mem_map is just set as node 0's */ |
c713216de [PATCH] Introduce... |
2374 |
if (pgdat == NODE_DATA(0)) { |
1da177e4c Linux-2.6.12-rc2 |
2375 |
mem_map = NODE_DATA(0)->node_mem_map; |
c713216de [PATCH] Introduce... |
2376 2377 2378 2379 2380 |
#ifdef CONFIG_ARCH_POPULATES_NODE_MAP if (page_to_pfn(mem_map) != pgdat->node_start_pfn) mem_map -= pgdat->node_start_pfn; #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */ } |
1da177e4c Linux-2.6.12-rc2 |
2381 |
#endif |
d41dee369 [PATCH] sparsemem... |
2382 |
#endif /* CONFIG_FLAT_NODE_MEM_MAP */ |
1da177e4c Linux-2.6.12-rc2 |
2383 |
} |
86356ab14 [PATCH] wait_tabl... |
2384 |
void __meminit free_area_init_node(int nid, struct pglist_data *pgdat, |
1da177e4c Linux-2.6.12-rc2 |
2385 2386 2387 2388 2389 |
unsigned long *zones_size, unsigned long node_start_pfn, unsigned long *zholes_size) { pgdat->node_id = nid; pgdat->node_start_pfn = node_start_pfn; |
c713216de [PATCH] Introduce... |
2390 |
calculate_node_totalpages(pgdat, zones_size, zholes_size); |
1da177e4c Linux-2.6.12-rc2 |
2391 2392 2393 2394 2395 |
alloc_node_mem_map(pgdat); free_area_init_core(pgdat, zones_size, zholes_size); } |
c713216de [PATCH] Introduce... |
2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 |
#ifdef CONFIG_ARCH_POPULATES_NODE_MAP /** * add_active_range - Register a range of PFNs backed by physical memory * @nid: The node ID the range resides on * @start_pfn: The start PFN of the available physical memory * @end_pfn: The end PFN of the available physical memory * * These ranges are stored in an early_node_map[] and later used by * free_area_init_nodes() to calculate zone sizes and holes. If the * range spans a memory hole, it is up to the architecture to ensure * the memory is not freed by the bootmem allocator. If possible * the range being registered will be merged with existing ranges. */ void __init add_active_range(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn) { int i; printk(KERN_DEBUG "Entering add_active_range(%d, %lu, %lu) " "%d entries of %d used ", nid, start_pfn, end_pfn, nr_nodemap_entries, MAX_ACTIVE_REGIONS); /* Merge with existing active regions if possible */ for (i = 0; i < nr_nodemap_entries; i++) { if (early_node_map[i].nid != nid) continue; /* Skip if an existing region covers this new one */ if (start_pfn >= early_node_map[i].start_pfn && end_pfn <= early_node_map[i].end_pfn) return; /* Merge forward if suitable */ if (start_pfn <= early_node_map[i].end_pfn && end_pfn > early_node_map[i].end_pfn) { early_node_map[i].end_pfn = end_pfn; return; } /* Merge backward if suitable */ if (start_pfn < early_node_map[i].end_pfn && end_pfn >= early_node_map[i].start_pfn) { early_node_map[i].start_pfn = start_pfn; return; } } /* Check that early_node_map is large enough */ if (i >= MAX_ACTIVE_REGIONS) { printk(KERN_CRIT "More than %d memory regions, truncating ", MAX_ACTIVE_REGIONS); return; } early_node_map[i].nid = nid; early_node_map[i].start_pfn = start_pfn; early_node_map[i].end_pfn = end_pfn; nr_nodemap_entries = i + 1; } /** * shrink_active_range - Shrink an existing registered range of PFNs * @nid: The node id the range is on that should be shrunk * @old_end_pfn: The old end PFN of the range * @new_end_pfn: The new PFN of the range * * i386 with NUMA use alloc_remap() to store a node_mem_map on a local node. * The map is kept at the end physical page range that has already been * registered with add_active_range(). This function allows an arch to shrink * an existing registered range. */ void __init shrink_active_range(unsigned int nid, unsigned long old_end_pfn, unsigned long new_end_pfn) { int i; /* Find the old active region end and shrink */ for_each_active_range_index_in_nid(i, nid) if (early_node_map[i].end_pfn == old_end_pfn) { early_node_map[i].end_pfn = new_end_pfn; break; } } /** * remove_all_active_ranges - Remove all currently registered regions |
88ca3b94e [PATCH] page_allo... |
2485 |
* |
c713216de [PATCH] Introduce... |
2486 2487 2488 2489 |
* During discovery, it may be found that a table like SRAT is invalid * and an alternative discovery method must be used. This function removes * all currently registered regions. */ |
88ca3b94e [PATCH] page_allo... |
2490 |
void __init remove_all_active_ranges(void) |
c713216de [PATCH] Introduce... |
2491 2492 2493 |
{ memset(early_node_map, 0, sizeof(early_node_map)); nr_nodemap_entries = 0; |
fb01439c5 [PATCH] Allow an ... |
2494 2495 2496 2497 |
#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE memset(node_boundary_start_pfn, 0, sizeof(node_boundary_start_pfn)); memset(node_boundary_end_pfn, 0, sizeof(node_boundary_end_pfn)); #endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */ |
c713216de [PATCH] Introduce... |
2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 |
} /* Compare two active node_active_regions */ static int __init cmp_node_active_region(const void *a, const void *b) { struct node_active_region *arange = (struct node_active_region *)a; struct node_active_region *brange = (struct node_active_region *)b; /* Done this way to avoid overflows */ if (arange->start_pfn > brange->start_pfn) return 1; if (arange->start_pfn < brange->start_pfn) return -1; return 0; } /* sort the node_map by start_pfn */ static void __init sort_node_map(void) { sort(early_node_map, (size_t)nr_nodemap_entries, sizeof(struct node_active_region), cmp_node_active_region, NULL); } /* Find the lowest pfn for a node. This depends on a sorted early_node_map */ unsigned long __init find_min_pfn_for_node(unsigned long nid) { int i; /* Assuming a sorted map, the first range found has the starting pfn */ for_each_active_range_index_in_nid(i, nid) return early_node_map[i].start_pfn; printk(KERN_WARNING "Could not find start_pfn for node %lu ", nid); return 0; } /** * 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... |
2541 |
* add_active_range(). |
c713216de [PATCH] Introduce... |
2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 |
*/ unsigned long __init find_min_pfn_with_active_regions(void) { return find_min_pfn_for_node(MAX_NUMNODES); } /** * find_max_pfn_with_active_regions - Find the maximum PFN registered * * It returns the maximum PFN based on information provided via |
88ca3b94e [PATCH] page_allo... |
2552 |
* add_active_range(). |
c713216de [PATCH] Introduce... |
2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 |
*/ unsigned long __init find_max_pfn_with_active_regions(void) { int i; unsigned long max_pfn = 0; for (i = 0; i < nr_nodemap_entries; i++) max_pfn = max(max_pfn, early_node_map[i].end_pfn); return max_pfn; } /** * free_area_init_nodes - Initialise all pg_data_t and zone data |
88ca3b94e [PATCH] page_allo... |
2567 |
* @max_zone_pfn: an array of max PFNs for each zone |
c713216de [PATCH] Introduce... |
2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 |
* * 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) { unsigned long nid; enum zone_type i; /* 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++) { 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]); } /* Regions in the early_node_map can be in any order */ sort_node_map(); /* Print out the zone ranges */ printk("Zone PFN ranges: "); for (i = 0; i < MAX_NR_ZONES; i++) printk(" %-8s %8lu -> %8lu ", zone_names[i], arch_zone_lowest_possible_pfn[i], arch_zone_highest_possible_pfn[i]); /* Print out the early_node_map[] */ printk("early_node_map[%d] active PFN ranges ", nr_nodemap_entries); for (i = 0; i < nr_nodemap_entries; i++) printk(" %3d: %8lu -> %8lu ", early_node_map[i].nid, early_node_map[i].start_pfn, early_node_map[i].end_pfn); /* Initialise every node */ for_each_online_node(nid) { pg_data_t *pgdat = NODE_DATA(nid); free_area_init_node(nid, pgdat, NULL, find_min_pfn_for_node(nid), NULL); } } #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */ |
0e0b864e0 [PATCH] Account f... |
2627 |
/** |
88ca3b94e [PATCH] page_allo... |
2628 2629 |
* 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... |
2630 2631 2632 2633 |
* * 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... |
2634 2635 2636 |
* 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... |
2637 2638 2639 2640 2641 |
*/ void __init set_dma_reserve(unsigned long new_dma_reserve) { dma_reserve = new_dma_reserve; } |
93b7504e3 [PATCH] Introduce... |
2642 |
#ifndef CONFIG_NEED_MULTIPLE_NODES |
1da177e4c Linux-2.6.12-rc2 |
2643 2644 2645 2646 |
static bootmem_data_t contig_bootmem_data; struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data }; EXPORT_SYMBOL(contig_page_data); |
93b7504e3 [PATCH] Introduce... |
2647 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
2648 2649 2650 |
void __init free_area_init(unsigned long *zones_size) { |
93b7504e3 [PATCH] Introduce... |
2651 |
free_area_init_node(0, NODE_DATA(0), zones_size, |
1da177e4c Linux-2.6.12-rc2 |
2652 2653 |
__pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL); } |
1da177e4c Linux-2.6.12-rc2 |
2654 |
|
1da177e4c Linux-2.6.12-rc2 |
2655 2656 2657 2658 2659 |
#ifdef CONFIG_HOTPLUG_CPU 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 |
2660 2661 |
if (action == CPU_DEAD) { |
1da177e4c Linux-2.6.12-rc2 |
2662 2663 |
local_irq_disable(); __drain_pages(cpu); |
f8891e5e1 [PATCH] Light wei... |
2664 |
vm_events_fold_cpu(cpu); |
1da177e4c Linux-2.6.12-rc2 |
2665 |
local_irq_enable(); |
2244b95a7 [PATCH] zoned vm ... |
2666 |
refresh_cpu_vm_stats(cpu); |
1da177e4c Linux-2.6.12-rc2 |
2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 |
} return NOTIFY_OK; } #endif /* CONFIG_HOTPLUG_CPU */ void __init page_alloc_init(void) { hotcpu_notifier(page_alloc_cpu_notify, 0); } /* |
cb45b0e96 [PATCH] overcommi... |
2678 2679 2680 2681 2682 2683 2684 |
* 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... |
2685 |
enum zone_type i, j; |
cb45b0e96 [PATCH] overcommi... |
2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 |
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]; } /* we treat pages_high as reserved pages. */ max += zone->pages_high; if (max > zone->present_pages) max = zone->present_pages; reserve_pages += max; } } totalreserve_pages = reserve_pages; } /* |
1da177e4c Linux-2.6.12-rc2 |
2710 2711 2712 2713 2714 2715 2716 2717 |
* 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... |
2718 |
enum zone_type j, idx; |
1da177e4c Linux-2.6.12-rc2 |
2719 |
|
ec936fc56 [PATCH] for_each_... |
2720 |
for_each_online_pgdat(pgdat) { |
1da177e4c Linux-2.6.12-rc2 |
2721 2722 2723 2724 2725 |
for (j = 0; j < MAX_NR_ZONES; j++) { struct zone *zone = pgdat->node_zones + j; unsigned long present_pages = zone->present_pages; zone->lowmem_reserve[j] = 0; |
2f6726e54 [PATCH] Apply typ... |
2726 2727 |
idx = j; while (idx) { |
1da177e4c Linux-2.6.12-rc2 |
2728 |
struct zone *lower_zone; |
2f6726e54 [PATCH] Apply typ... |
2729 |
idx--; |
1da177e4c Linux-2.6.12-rc2 |
2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 |
if (sysctl_lowmem_reserve_ratio[idx] < 1) sysctl_lowmem_reserve_ratio[idx] = 1; lower_zone = pgdat->node_zones + idx; lower_zone->lowmem_reserve[j] = present_pages / sysctl_lowmem_reserve_ratio[idx]; present_pages += lower_zone->present_pages; } } } |
cb45b0e96 [PATCH] overcommi... |
2740 2741 2742 |
/* update totalreserve_pages */ calculate_totalreserve_pages(); |
1da177e4c Linux-2.6.12-rc2 |
2743 |
} |
88ca3b94e [PATCH] page_allo... |
2744 2745 2746 2747 2748 |
/** * setup_per_zone_pages_min - called when min_free_kbytes changes. * * Ensures that the pages_{min,low,high} values for each zone are set correctly * with respect to min_free_kbytes. |
1da177e4c Linux-2.6.12-rc2 |
2749 |
*/ |
3947be196 [PATCH] memory ho... |
2750 |
void setup_per_zone_pages_min(void) |
1da177e4c Linux-2.6.12-rc2 |
2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 |
{ 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)) lowmem_pages += zone->present_pages; } for_each_zone(zone) { |
ac924c603 [PATCH] setup_per... |
2764 |
u64 tmp; |
1da177e4c Linux-2.6.12-rc2 |
2765 |
spin_lock_irqsave(&zone->lru_lock, flags); |
ac924c603 [PATCH] setup_per... |
2766 2767 |
tmp = (u64)pages_min * zone->present_pages; do_div(tmp, lowmem_pages); |
1da177e4c Linux-2.6.12-rc2 |
2768 2769 |
if (is_highmem(zone)) { /* |
669ed1752 [PATCH] mm: highm... |
2770 2771 2772 2773 2774 2775 2776 |
* __GFP_HIGH and PF_MEMALLOC allocations usually don't * need highmem pages, so cap pages_min to a small * value here. * * The (pages_high-pages_low) and (pages_low-pages_min) * deltas controls asynch page reclaim, and so should * not be capped for highmem. |
1da177e4c Linux-2.6.12-rc2 |
2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 |
*/ int min_pages; min_pages = zone->present_pages / 1024; if (min_pages < SWAP_CLUSTER_MAX) min_pages = SWAP_CLUSTER_MAX; if (min_pages > 128) min_pages = 128; zone->pages_min = min_pages; } else { |
669ed1752 [PATCH] mm: highm... |
2787 2788 |
/* * If it's a lowmem zone, reserve a number of pages |
1da177e4c Linux-2.6.12-rc2 |
2789 2790 |
* proportionate to the zone's size. */ |
669ed1752 [PATCH] mm: highm... |
2791 |
zone->pages_min = tmp; |
1da177e4c Linux-2.6.12-rc2 |
2792 |
} |
ac924c603 [PATCH] setup_per... |
2793 2794 |
zone->pages_low = zone->pages_min + (tmp >> 2); zone->pages_high = zone->pages_min + (tmp >> 1); |
1da177e4c Linux-2.6.12-rc2 |
2795 2796 |
spin_unlock_irqrestore(&zone->lru_lock, flags); } |
cb45b0e96 [PATCH] overcommi... |
2797 2798 2799 |
/* update totalreserve_pages */ calculate_totalreserve_pages(); |
1da177e4c Linux-2.6.12-rc2 |
2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 |
} /* * 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 */ static int __init init_per_zone_pages_min(void) { 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; setup_per_zone_pages_min(); setup_per_zone_lowmem_reserve(); return 0; } module_init(init_per_zone_pages_min) /* * 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, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) { proc_dointvec(table, write, file, buffer, length, ppos); setup_per_zone_pages_min(); return 0; } |
9614634fe [PATCH] ZVC/zone_... |
2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 |
#ifdef CONFIG_NUMA int sysctl_min_unmapped_ratio_sysctl_handler(ctl_table *table, int write, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) { struct zone *zone; int rc; rc = proc_dointvec_minmax(table, write, file, buffer, length, ppos); if (rc) return rc; for_each_zone(zone) |
8417bba4b [PATCH] Replace m... |
2867 |
zone->min_unmapped_pages = (zone->present_pages * |
9614634fe [PATCH] ZVC/zone_... |
2868 2869 2870 |
sysctl_min_unmapped_ratio) / 100; return 0; } |
0ff38490c [PATCH] zone_recl... |
2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 |
int sysctl_min_slab_ratio_sysctl_handler(ctl_table *table, int write, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) { struct zone *zone; int rc; rc = proc_dointvec_minmax(table, write, file, buffer, length, ppos); if (rc) return rc; for_each_zone(zone) zone->min_slab_pages = (zone->present_pages * sysctl_min_slab_ratio) / 100; return 0; } |
9614634fe [PATCH] ZVC/zone_... |
2887 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 |
/* * 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 * pages_min watermarks. The lowmem reserve ratio can only make sense * if in function of the boot time zone sizes. */ int lowmem_reserve_ratio_sysctl_handler(ctl_table *table, int write, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) { proc_dointvec_minmax(table, write, file, buffer, length, ppos); setup_per_zone_lowmem_reserve(); return 0; } |
8ad4b1fb8 [PATCH] Make high... |
2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 |
/* * 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, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) { struct zone *zone; unsigned int cpu; int ret; ret = proc_dointvec_minmax(table, write, file, buffer, length, ppos); if (!write || (ret == -EINVAL)) return ret; for_each_zone(zone) { for_each_online_cpu(cpu) { unsigned long high; high = zone->present_pages / percpu_pagelist_fraction; setup_pagelist_highmark(zone_pcp(zone, cpu), high); } } return 0; } |
f034b5d4e [XFRM]: Dynamic x... |
2929 |
int hashdist = HASHDIST_DEFAULT; |
1da177e4c Linux-2.6.12-rc2 |
2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 |
#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, unsigned long limit) { unsigned long long max = limit; 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 */ numentries = (flags & HASH_HIGHMEM) ? nr_all_pages : nr_kernel_pages; 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); } |
6e692ed37 [PATCH] fix alloc... |
2975 |
numentries = roundup_pow_of_two(numentries); |
1da177e4c Linux-2.6.12-rc2 |
2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 |
/* 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); } if (numentries > max) numentries = max; log2qty = long_log2(numentries); do { size = bucketsize << log2qty; if (flags & HASH_EARLY) table = alloc_bootmem(size); else if (hashdist) table = __vmalloc(size, GFP_ATOMIC, PAGE_KERNEL); else { unsigned long order; for (order = 0; ((1UL << order) << PAGE_SHIFT) < size; order++) ; table = (void*) __get_free_pages(GFP_ATOMIC, order); } } while (!table && size > PAGE_SIZE && --log2qty); if (!table) panic("Failed to allocate %s hash table ", tablename); printk("%s hash table entries: %d (order: %d, %lu bytes) ", tablename, (1U << log2qty), long_log2(size) - PAGE_SHIFT, size); if (_hash_shift) *_hash_shift = log2qty; if (_hash_mask) *_hash_mask = (1 << log2qty) - 1; return table; } |
a117e66ed [PATCH] unify pfn... |
3020 3021 |
#ifdef CONFIG_OUT_OF_LINE_PFN_TO_PAGE |
a117e66ed [PATCH] unify pfn... |
3022 3023 |
struct page *pfn_to_page(unsigned long pfn) { |
67de64821 [PATCH] squash du... |
3024 |
return __pfn_to_page(pfn); |
a117e66ed [PATCH] unify pfn... |
3025 3026 3027 |
} unsigned long page_to_pfn(struct page *page) { |
67de64821 [PATCH] squash du... |
3028 |
return __page_to_pfn(page); |
a117e66ed [PATCH] unify pfn... |
3029 |
} |
a117e66ed [PATCH] unify pfn... |
3030 3031 3032 |
EXPORT_SYMBOL(pfn_to_page); EXPORT_SYMBOL(page_to_pfn); #endif /* CONFIG_OUT_OF_LINE_PFN_TO_PAGE */ |