Blame view
mm/vmstat.c
53.1 KB
457c89965
|
1 |
// SPDX-License-Identifier: GPL-2.0-only |
f6ac2354d
|
2 3 4 5 6 |
/* * linux/mm/vmstat.c * * Manages VM statistics * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds |
2244b95a7
|
7 8 9 10 |
* * zoned VM statistics * Copyright (C) 2006 Silicon Graphics, Inc., * Christoph Lameter <christoph@lameter.com> |
7cc36bbdd
|
11 |
* Copyright (C) 2008-2014 Christoph Lameter |
f6ac2354d
|
12 |
*/ |
8f32f7e5a
|
13 |
#include <linux/fs.h> |
f6ac2354d
|
14 |
#include <linux/mm.h> |
4e950f6f0
|
15 |
#include <linux/err.h> |
2244b95a7
|
16 |
#include <linux/module.h> |
5a0e3ad6a
|
17 |
#include <linux/slab.h> |
df9ecaba3
|
18 |
#include <linux/cpu.h> |
7cc36bbdd
|
19 |
#include <linux/cpumask.h> |
c748e1340
|
20 |
#include <linux/vmstat.h> |
3c4868710
|
21 22 23 |
#include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/debugfs.h> |
e8edc6e03
|
24 |
#include <linux/sched.h> |
f1a5ab121
|
25 |
#include <linux/math64.h> |
79da826ae
|
26 |
#include <linux/writeback.h> |
36deb0be3
|
27 |
#include <linux/compaction.h> |
6e543d578
|
28 |
#include <linux/mm_inline.h> |
48c96a368
|
29 30 |
#include <linux/page_ext.h> #include <linux/page_owner.h> |
6e543d578
|
31 32 |
#include "internal.h" |
f6ac2354d
|
33 |
|
1d90ca897
|
34 |
#define NUMA_STATS_THRESHOLD (U16_MAX - 2) |
4518085e1
|
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 |
#ifdef CONFIG_NUMA int sysctl_vm_numa_stat = ENABLE_NUMA_STAT; /* zero numa counters within a zone */ static void zero_zone_numa_counters(struct zone *zone) { int item, cpu; for (item = 0; item < NR_VM_NUMA_STAT_ITEMS; item++) { atomic_long_set(&zone->vm_numa_stat[item], 0); for_each_online_cpu(cpu) per_cpu_ptr(zone->pageset, cpu)->vm_numa_stat_diff[item] = 0; } } /* zero numa counters of all the populated zones */ static void zero_zones_numa_counters(void) { struct zone *zone; for_each_populated_zone(zone) zero_zone_numa_counters(zone); } /* zero global numa counters */ static void zero_global_numa_counters(void) { int item; for (item = 0; item < NR_VM_NUMA_STAT_ITEMS; item++) atomic_long_set(&vm_numa_stat[item], 0); } static void invalid_numa_statistics(void) { zero_zones_numa_counters(); zero_global_numa_counters(); } static DEFINE_MUTEX(vm_numa_stat_lock); int sysctl_vm_numa_stat_handler(struct ctl_table *table, int write, |
32927393d
|
78 |
void *buffer, size_t *length, loff_t *ppos) |
4518085e1
|
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 |
{ int ret, oldval; mutex_lock(&vm_numa_stat_lock); if (write) oldval = sysctl_vm_numa_stat; ret = proc_dointvec_minmax(table, write, buffer, length, ppos); if (ret || !write) goto out; if (oldval == sysctl_vm_numa_stat) goto out; else if (sysctl_vm_numa_stat == ENABLE_NUMA_STAT) { static_branch_enable(&vm_numa_stat_key); pr_info("enable numa statistics "); } else { static_branch_disable(&vm_numa_stat_key); invalid_numa_statistics(); pr_info("disable numa statistics, and clear numa counters "); } out: mutex_unlock(&vm_numa_stat_lock); return ret; } #endif |
f8891e5e1
|
107 108 109 |
#ifdef CONFIG_VM_EVENT_COUNTERS DEFINE_PER_CPU(struct vm_event_state, vm_event_states) = {{0}}; EXPORT_PER_CPU_SYMBOL(vm_event_states); |
31f961a89
|
110 |
static void sum_vm_events(unsigned long *ret) |
f8891e5e1
|
111 |
{ |
9eccf2a81
|
112 |
int cpu; |
f8891e5e1
|
113 114 115 |
int i; memset(ret, 0, NR_VM_EVENT_ITEMS * sizeof(unsigned long)); |
31f961a89
|
116 |
for_each_online_cpu(cpu) { |
f8891e5e1
|
117 |
struct vm_event_state *this = &per_cpu(vm_event_states, cpu); |
f8891e5e1
|
118 119 120 121 122 123 124 125 126 127 128 129 |
for (i = 0; i < NR_VM_EVENT_ITEMS; i++) ret[i] += this->event[i]; } } /* * Accumulate the vm event counters across all CPUs. * The result is unavoidably approximate - it can change * during and after execution of this function. */ void all_vm_events(unsigned long *ret) { |
b5be11329
|
130 |
get_online_cpus(); |
31f961a89
|
131 |
sum_vm_events(ret); |
b5be11329
|
132 |
put_online_cpus(); |
f8891e5e1
|
133 |
} |
32dd66fce
|
134 |
EXPORT_SYMBOL_GPL(all_vm_events); |
f8891e5e1
|
135 |
|
f8891e5e1
|
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 |
/* * Fold the foreign cpu events into our own. * * This is adding to the events on one processor * but keeps the global counts constant. */ void vm_events_fold_cpu(int cpu) { struct vm_event_state *fold_state = &per_cpu(vm_event_states, cpu); int i; for (i = 0; i < NR_VM_EVENT_ITEMS; i++) { count_vm_events(i, fold_state->event[i]); fold_state->event[i] = 0; } } |
f8891e5e1
|
152 153 |
#endif /* CONFIG_VM_EVENT_COUNTERS */ |
2244b95a7
|
154 155 156 157 158 |
/* * Manage combined zone based / global counters * * vm_stat contains the global counters */ |
75ef71840
|
159 |
atomic_long_t vm_zone_stat[NR_VM_ZONE_STAT_ITEMS] __cacheline_aligned_in_smp; |
3a321d2a3
|
160 |
atomic_long_t vm_numa_stat[NR_VM_NUMA_STAT_ITEMS] __cacheline_aligned_in_smp; |
75ef71840
|
161 162 |
atomic_long_t vm_node_stat[NR_VM_NODE_STAT_ITEMS] __cacheline_aligned_in_smp; EXPORT_SYMBOL(vm_zone_stat); |
3a321d2a3
|
163 |
EXPORT_SYMBOL(vm_numa_stat); |
75ef71840
|
164 |
EXPORT_SYMBOL(vm_node_stat); |
2244b95a7
|
165 166 |
#ifdef CONFIG_SMP |
b44129b30
|
167 |
int calculate_pressure_threshold(struct zone *zone) |
88f5acf88
|
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 |
{ int threshold; int watermark_distance; /* * As vmstats are not up to date, there is drift between the estimated * and real values. For high thresholds and a high number of CPUs, it * is possible for the min watermark to be breached while the estimated * value looks fine. The pressure threshold is a reduced value such * that even the maximum amount of drift will not accidentally breach * the min watermark */ watermark_distance = low_wmark_pages(zone) - min_wmark_pages(zone); threshold = max(1, (int)(watermark_distance / num_online_cpus())); /* * Maximum threshold is 125 */ threshold = min(125, threshold); return threshold; } |
b44129b30
|
190 |
int calculate_normal_threshold(struct zone *zone) |
df9ecaba3
|
191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 |
{ int threshold; int mem; /* memory in 128 MB units */ /* * The threshold scales with the number of processors and the amount * of memory per zone. More memory means that we can defer updates for * longer, more processors could lead to more contention. * fls() is used to have a cheap way of logarithmic scaling. * * Some sample thresholds: * * Threshold Processors (fls) Zonesize fls(mem+1) * ------------------------------------------------------------------ * 8 1 1 0.9-1 GB 4 * 16 2 2 0.9-1 GB 4 * 20 2 2 1-2 GB 5 * 24 2 2 2-4 GB 6 * 28 2 2 4-8 GB 7 * 32 2 2 8-16 GB 8 * 4 2 2 <128M 1 * 30 4 3 2-4 GB 5 * 48 4 3 8-16 GB 8 * 32 8 4 1-2 GB 4 * 32 8 4 0.9-1GB 4 * 10 16 5 <128M 1 * 40 16 5 900M 4 * 70 64 7 2-4 GB 5 * 84 64 7 4-8 GB 6 * 108 512 9 4-8 GB 6 * 125 1024 10 8-16 GB 8 * 125 1024 10 16-32 GB 9 */ |
9705bea5f
|
224 |
mem = zone_managed_pages(zone) >> (27 - PAGE_SHIFT); |
df9ecaba3
|
225 226 227 228 229 230 231 232 233 234 |
threshold = 2 * fls(num_online_cpus()) * (1 + fls(mem)); /* * Maximum threshold is 125 */ threshold = min(125, threshold); return threshold; } |
2244b95a7
|
235 236 |
/* |
df9ecaba3
|
237 |
* Refresh the thresholds for each zone. |
2244b95a7
|
238 |
*/ |
a6cccdc36
|
239 |
void refresh_zone_stat_thresholds(void) |
2244b95a7
|
240 |
{ |
75ef71840
|
241 |
struct pglist_data *pgdat; |
df9ecaba3
|
242 243 244 |
struct zone *zone; int cpu; int threshold; |
75ef71840
|
245 246 247 248 249 250 |
/* Zero current pgdat thresholds */ for_each_online_pgdat(pgdat) { for_each_online_cpu(cpu) { per_cpu_ptr(pgdat->per_cpu_nodestats, cpu)->stat_threshold = 0; } } |
ee99c71c5
|
251 |
for_each_populated_zone(zone) { |
75ef71840
|
252 |
struct pglist_data *pgdat = zone->zone_pgdat; |
aa4548403
|
253 |
unsigned long max_drift, tolerate_drift; |
b44129b30
|
254 |
threshold = calculate_normal_threshold(zone); |
df9ecaba3
|
255 |
|
75ef71840
|
256 257 |
for_each_online_cpu(cpu) { int pgdat_threshold; |
99dcc3e5a
|
258 259 |
per_cpu_ptr(zone->pageset, cpu)->stat_threshold = threshold; |
1d90ca897
|
260 |
|
75ef71840
|
261 262 263 264 265 |
/* Base nodestat threshold on the largest populated zone. */ pgdat_threshold = per_cpu_ptr(pgdat->per_cpu_nodestats, cpu)->stat_threshold; per_cpu_ptr(pgdat->per_cpu_nodestats, cpu)->stat_threshold = max(threshold, pgdat_threshold); } |
aa4548403
|
266 267 268 269 270 271 272 273 274 275 |
/* * Only set percpu_drift_mark if there is a danger that * NR_FREE_PAGES reports the low watermark is ok when in fact * the min watermark could be breached by an allocation */ tolerate_drift = low_wmark_pages(zone) - min_wmark_pages(zone); max_drift = num_online_cpus() * threshold; if (max_drift > tolerate_drift) zone->percpu_drift_mark = high_wmark_pages(zone) + max_drift; |
df9ecaba3
|
276 |
} |
2244b95a7
|
277 |
} |
b44129b30
|
278 279 |
void set_pgdat_percpu_threshold(pg_data_t *pgdat, int (*calculate_pressure)(struct zone *)) |
88f5acf88
|
280 281 282 283 284 |
{ struct zone *zone; int cpu; int threshold; int i; |
88f5acf88
|
285 286 287 288 |
for (i = 0; i < pgdat->nr_zones; i++) { zone = &pgdat->node_zones[i]; if (!zone->percpu_drift_mark) continue; |
b44129b30
|
289 |
threshold = (*calculate_pressure)(zone); |
1d90ca897
|
290 |
for_each_online_cpu(cpu) |
88f5acf88
|
291 292 293 |
per_cpu_ptr(zone->pageset, cpu)->stat_threshold = threshold; } |
88f5acf88
|
294 |
} |
2244b95a7
|
295 |
/* |
bea04b073
|
296 297 298 |
* For use when we know that interrupts are disabled, * or when we know that preemption is disabled and that * particular counter cannot be updated from interrupt context. |
2244b95a7
|
299 300 |
*/ void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item, |
6cdb18ad9
|
301 |
long delta) |
2244b95a7
|
302 |
{ |
12938a922
|
303 304 |
struct per_cpu_pageset __percpu *pcp = zone->pageset; s8 __percpu *p = pcp->vm_stat_diff + item; |
2244b95a7
|
305 |
long x; |
12938a922
|
306 307 308 |
long t; x = delta + __this_cpu_read(*p); |
2244b95a7
|
309 |
|
12938a922
|
310 |
t = __this_cpu_read(pcp->stat_threshold); |
2244b95a7
|
311 |
|
12938a922
|
312 |
if (unlikely(x > t || x < -t)) { |
2244b95a7
|
313 314 315 |
zone_page_state_add(x, zone, item); x = 0; } |
12938a922
|
316 |
__this_cpu_write(*p, x); |
2244b95a7
|
317 318 |
} EXPORT_SYMBOL(__mod_zone_page_state); |
75ef71840
|
319 320 321 322 323 324 325 |
void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item, long delta) { struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats; s8 __percpu *p = pcp->vm_node_stat_diff + item; long x; long t; |
ea426c2a7
|
326 327 328 329 |
if (vmstat_item_in_bytes(item)) { VM_WARN_ON_ONCE(delta & (PAGE_SIZE - 1)); delta >>= PAGE_SHIFT; } |
75ef71840
|
330 331 332 333 334 335 336 337 338 339 340 |
x = delta + __this_cpu_read(*p); t = __this_cpu_read(pcp->stat_threshold); if (unlikely(x > t || x < -t)) { node_page_state_add(x, pgdat, item); x = 0; } __this_cpu_write(*p, x); } EXPORT_SYMBOL(__mod_node_page_state); |
2244b95a7
|
341 |
/* |
2244b95a7
|
342 343 344 345 346 347 348 349 350 |
* Optimized increment and decrement functions. * * These are only for a single page and therefore can take a struct page * * argument instead of struct zone *. This allows the inclusion of the code * generated for page_zone(page) into the optimized functions. * * No overflow check is necessary and therefore the differential can be * incremented or decremented in place which may allow the compilers to * generate better code. |
2244b95a7
|
351 352 353 |
* The increment or decrement is known and therefore one boundary check can * be omitted. * |
df9ecaba3
|
354 355 356 |
* NOTE: These functions are very performance sensitive. Change only * with care. * |
2244b95a7
|
357 358 359 360 361 362 363 |
* Some processors have inc/dec instructions that are atomic vs an interrupt. * However, the code must first determine the differential location in a zone * based on the processor number and then inc/dec the counter. There is no * guarantee without disabling preemption that the processor will not change * in between and therefore the atomicity vs. interrupt cannot be exploited * in a useful way here. */ |
c87853859
|
364 |
void __inc_zone_state(struct zone *zone, enum zone_stat_item item) |
2244b95a7
|
365 |
{ |
12938a922
|
366 367 368 |
struct per_cpu_pageset __percpu *pcp = zone->pageset; s8 __percpu *p = pcp->vm_stat_diff + item; s8 v, t; |
2244b95a7
|
369 |
|
908ee0f12
|
370 |
v = __this_cpu_inc_return(*p); |
12938a922
|
371 372 373 |
t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v > t)) { s8 overstep = t >> 1; |
df9ecaba3
|
374 |
|
12938a922
|
375 376 |
zone_page_state_add(v + overstep, zone, item); __this_cpu_write(*p, -overstep); |
2244b95a7
|
377 378 |
} } |
ca889e6c4
|
379 |
|
75ef71840
|
380 381 382 383 384 |
void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item) { struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats; s8 __percpu *p = pcp->vm_node_stat_diff + item; s8 v, t; |
ea426c2a7
|
385 |
VM_WARN_ON_ONCE(vmstat_item_in_bytes(item)); |
75ef71840
|
386 387 388 389 390 391 392 393 394 |
v = __this_cpu_inc_return(*p); t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v > t)) { s8 overstep = t >> 1; node_page_state_add(v + overstep, pgdat, item); __this_cpu_write(*p, -overstep); } } |
ca889e6c4
|
395 396 397 398 |
void __inc_zone_page_state(struct page *page, enum zone_stat_item item) { __inc_zone_state(page_zone(page), item); } |
2244b95a7
|
399 |
EXPORT_SYMBOL(__inc_zone_page_state); |
75ef71840
|
400 401 402 403 404 |
void __inc_node_page_state(struct page *page, enum node_stat_item item) { __inc_node_state(page_pgdat(page), item); } EXPORT_SYMBOL(__inc_node_page_state); |
c87853859
|
405 |
void __dec_zone_state(struct zone *zone, enum zone_stat_item item) |
2244b95a7
|
406 |
{ |
12938a922
|
407 408 409 |
struct per_cpu_pageset __percpu *pcp = zone->pageset; s8 __percpu *p = pcp->vm_stat_diff + item; s8 v, t; |
2244b95a7
|
410 |
|
908ee0f12
|
411 |
v = __this_cpu_dec_return(*p); |
12938a922
|
412 413 414 |
t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v < - t)) { s8 overstep = t >> 1; |
2244b95a7
|
415 |
|
12938a922
|
416 417 |
zone_page_state_add(v - overstep, zone, item); __this_cpu_write(*p, overstep); |
2244b95a7
|
418 419 |
} } |
c87853859
|
420 |
|
75ef71840
|
421 422 423 424 425 |
void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item) { struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats; s8 __percpu *p = pcp->vm_node_stat_diff + item; s8 v, t; |
ea426c2a7
|
426 |
VM_WARN_ON_ONCE(vmstat_item_in_bytes(item)); |
75ef71840
|
427 428 429 430 431 432 433 434 435 |
v = __this_cpu_dec_return(*p); t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v < - t)) { s8 overstep = t >> 1; node_page_state_add(v - overstep, pgdat, item); __this_cpu_write(*p, overstep); } } |
c87853859
|
436 437 438 439 |
void __dec_zone_page_state(struct page *page, enum zone_stat_item item) { __dec_zone_state(page_zone(page), item); } |
2244b95a7
|
440 |
EXPORT_SYMBOL(__dec_zone_page_state); |
75ef71840
|
441 442 443 444 445 |
void __dec_node_page_state(struct page *page, enum node_stat_item item) { __dec_node_state(page_pgdat(page), item); } EXPORT_SYMBOL(__dec_node_page_state); |
4156153c4
|
446 |
#ifdef CONFIG_HAVE_CMPXCHG_LOCAL |
7c8391206
|
447 448 449 450 451 452 453 454 455 456 457 458 |
/* * If we have cmpxchg_local support then we do not need to incur the overhead * that comes with local_irq_save/restore if we use this_cpu_cmpxchg. * * mod_state() modifies the zone counter state through atomic per cpu * operations. * * Overstep mode specifies how overstep should handled: * 0 No overstepping * 1 Overstepping half of threshold * -1 Overstepping minus half of threshold */ |
75ef71840
|
459 460 |
static inline void mod_zone_state(struct zone *zone, enum zone_stat_item item, long delta, int overstep_mode) |
7c8391206
|
461 462 463 464 465 466 467 468 469 470 471 |
{ struct per_cpu_pageset __percpu *pcp = zone->pageset; s8 __percpu *p = pcp->vm_stat_diff + item; long o, n, t, z; do { z = 0; /* overflow to zone counters */ /* * The fetching of the stat_threshold is racy. We may apply * a counter threshold to the wrong the cpu if we get |
d3bc23671
|
472 473 474 475 476 477 |
* rescheduled while executing here. However, the next * counter update will apply the threshold again and * therefore bring the counter under the threshold again. * * Most of the time the thresholds are the same anyways * for all cpus in a zone. |
7c8391206
|
478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 |
*/ t = this_cpu_read(pcp->stat_threshold); o = this_cpu_read(*p); n = delta + o; if (n > t || n < -t) { int os = overstep_mode * (t >> 1) ; /* Overflow must be added to zone counters */ z = n + os; n = -os; } } while (this_cpu_cmpxchg(*p, o, n) != o); if (z) zone_page_state_add(z, zone, item); } void mod_zone_page_state(struct zone *zone, enum zone_stat_item item, |
6cdb18ad9
|
498 |
long delta) |
7c8391206
|
499 |
{ |
75ef71840
|
500 |
mod_zone_state(zone, item, delta, 0); |
7c8391206
|
501 502 |
} EXPORT_SYMBOL(mod_zone_page_state); |
7c8391206
|
503 504 |
void inc_zone_page_state(struct page *page, enum zone_stat_item item) { |
75ef71840
|
505 |
mod_zone_state(page_zone(page), item, 1, 1); |
7c8391206
|
506 507 508 509 510 |
} EXPORT_SYMBOL(inc_zone_page_state); void dec_zone_page_state(struct page *page, enum zone_stat_item item) { |
75ef71840
|
511 |
mod_zone_state(page_zone(page), item, -1, -1); |
7c8391206
|
512 513 |
} EXPORT_SYMBOL(dec_zone_page_state); |
75ef71840
|
514 515 516 517 518 519 520 |
static inline void mod_node_state(struct pglist_data *pgdat, enum node_stat_item item, int delta, int overstep_mode) { struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats; s8 __percpu *p = pcp->vm_node_stat_diff + item; long o, n, t, z; |
ea426c2a7
|
521 522 523 524 |
if (vmstat_item_in_bytes(item)) { VM_WARN_ON_ONCE(delta & (PAGE_SIZE - 1)); delta >>= PAGE_SHIFT; } |
75ef71840
|
525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 |
do { z = 0; /* overflow to node counters */ /* * The fetching of the stat_threshold is racy. We may apply * a counter threshold to the wrong the cpu if we get * rescheduled while executing here. However, the next * counter update will apply the threshold again and * therefore bring the counter under the threshold again. * * Most of the time the thresholds are the same anyways * for all cpus in a node. */ t = this_cpu_read(pcp->stat_threshold); o = this_cpu_read(*p); n = delta + o; if (n > t || n < -t) { int os = overstep_mode * (t >> 1) ; /* Overflow must be added to node counters */ z = n + os; n = -os; } } while (this_cpu_cmpxchg(*p, o, n) != o); if (z) node_page_state_add(z, pgdat, item); } void mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item, long delta) { mod_node_state(pgdat, item, delta, 0); } EXPORT_SYMBOL(mod_node_page_state); void inc_node_state(struct pglist_data *pgdat, enum node_stat_item item) { mod_node_state(pgdat, item, 1, 1); } void inc_node_page_state(struct page *page, enum node_stat_item item) { mod_node_state(page_pgdat(page), item, 1, 1); } EXPORT_SYMBOL(inc_node_page_state); void dec_node_page_state(struct page *page, enum node_stat_item item) { mod_node_state(page_pgdat(page), item, -1, -1); } EXPORT_SYMBOL(dec_node_page_state); |
7c8391206
|
579 580 581 582 583 |
#else /* * Use interrupt disable to serialize counter updates */ void mod_zone_page_state(struct zone *zone, enum zone_stat_item item, |
6cdb18ad9
|
584 |
long delta) |
7c8391206
|
585 586 587 588 589 590 591 592 |
{ unsigned long flags; local_irq_save(flags); __mod_zone_page_state(zone, item, delta); local_irq_restore(flags); } EXPORT_SYMBOL(mod_zone_page_state); |
2244b95a7
|
593 594 595 596 |
void inc_zone_page_state(struct page *page, enum zone_stat_item item) { unsigned long flags; struct zone *zone; |
2244b95a7
|
597 598 599 |
zone = page_zone(page); local_irq_save(flags); |
ca889e6c4
|
600 |
__inc_zone_state(zone, item); |
2244b95a7
|
601 602 603 604 605 606 607 |
local_irq_restore(flags); } EXPORT_SYMBOL(inc_zone_page_state); void dec_zone_page_state(struct page *page, enum zone_stat_item item) { unsigned long flags; |
2244b95a7
|
608 |
|
2244b95a7
|
609 |
local_irq_save(flags); |
a302eb4e4
|
610 |
__dec_zone_page_state(page, item); |
2244b95a7
|
611 612 613 |
local_irq_restore(flags); } EXPORT_SYMBOL(dec_zone_page_state); |
75ef71840
|
614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 |
void inc_node_state(struct pglist_data *pgdat, enum node_stat_item item) { unsigned long flags; local_irq_save(flags); __inc_node_state(pgdat, item); local_irq_restore(flags); } EXPORT_SYMBOL(inc_node_state); void mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item, long delta) { unsigned long flags; local_irq_save(flags); __mod_node_page_state(pgdat, item, delta); local_irq_restore(flags); } EXPORT_SYMBOL(mod_node_page_state); void inc_node_page_state(struct page *page, enum node_stat_item item) { unsigned long flags; struct pglist_data *pgdat; pgdat = page_pgdat(page); local_irq_save(flags); __inc_node_state(pgdat, item); local_irq_restore(flags); } EXPORT_SYMBOL(inc_node_page_state); void dec_node_page_state(struct page *page, enum node_stat_item item) { unsigned long flags; local_irq_save(flags); __dec_node_page_state(page, item); local_irq_restore(flags); } EXPORT_SYMBOL(dec_node_page_state); #endif |
7cc36bbdd
|
657 658 659 660 661 |
/* * Fold a differential into the global counters. * Returns the number of counters updated. */ |
3a321d2a3
|
662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 |
#ifdef CONFIG_NUMA static int fold_diff(int *zone_diff, int *numa_diff, int *node_diff) { int i; int changes = 0; for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) if (zone_diff[i]) { atomic_long_add(zone_diff[i], &vm_zone_stat[i]); changes++; } for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) if (numa_diff[i]) { atomic_long_add(numa_diff[i], &vm_numa_stat[i]); changes++; } for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) if (node_diff[i]) { atomic_long_add(node_diff[i], &vm_node_stat[i]); changes++; } return changes; } #else |
75ef71840
|
688 |
static int fold_diff(int *zone_diff, int *node_diff) |
4edb0748b
|
689 690 |
{ int i; |
7cc36bbdd
|
691 |
int changes = 0; |
4edb0748b
|
692 693 |
for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) |
75ef71840
|
694 695 696 697 698 699 700 701 |
if (zone_diff[i]) { atomic_long_add(zone_diff[i], &vm_zone_stat[i]); changes++; } for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) if (node_diff[i]) { atomic_long_add(node_diff[i], &vm_node_stat[i]); |
7cc36bbdd
|
702 703 704 |
changes++; } return changes; |
4edb0748b
|
705 |
} |
3a321d2a3
|
706 |
#endif /* CONFIG_NUMA */ |
4edb0748b
|
707 |
|
2244b95a7
|
708 |
/* |
2bb921e52
|
709 |
* Update the zone counters for the current cpu. |
a7f75e258
|
710 |
* |
4037d4522
|
711 712 713 714 715 716 717 718 719 720 |
* Note that refresh_cpu_vm_stats strives to only access * node local memory. The per cpu pagesets on remote zones are placed * in the memory local to the processor using that pageset. So the * loop over all zones will access a series of cachelines local to * the processor. * * The call to zone_page_state_add updates the cachelines with the * statistics in the remote zone struct as well as the global cachelines * with the global counters. These could cause remote node cache line * bouncing and will have to be only done when necessary. |
7cc36bbdd
|
721 722 |
* * The function returns the number of global counters updated. |
2244b95a7
|
723 |
*/ |
0eb77e988
|
724 |
static int refresh_cpu_vm_stats(bool do_pagesets) |
2244b95a7
|
725 |
{ |
75ef71840
|
726 |
struct pglist_data *pgdat; |
2244b95a7
|
727 728 |
struct zone *zone; int i; |
75ef71840
|
729 |
int global_zone_diff[NR_VM_ZONE_STAT_ITEMS] = { 0, }; |
3a321d2a3
|
730 731 732 |
#ifdef CONFIG_NUMA int global_numa_diff[NR_VM_NUMA_STAT_ITEMS] = { 0, }; #endif |
75ef71840
|
733 |
int global_node_diff[NR_VM_NODE_STAT_ITEMS] = { 0, }; |
7cc36bbdd
|
734 |
int changes = 0; |
2244b95a7
|
735 |
|
ee99c71c5
|
736 |
for_each_populated_zone(zone) { |
fbc2edb05
|
737 |
struct per_cpu_pageset __percpu *p = zone->pageset; |
2244b95a7
|
738 |
|
fbc2edb05
|
739 740 |
for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) { int v; |
2244b95a7
|
741 |
|
fbc2edb05
|
742 743 |
v = this_cpu_xchg(p->vm_stat_diff[i], 0); if (v) { |
a7f75e258
|
744 |
|
a7f75e258
|
745 |
atomic_long_add(v, &zone->vm_stat[i]); |
75ef71840
|
746 |
global_zone_diff[i] += v; |
4037d4522
|
747 748 |
#ifdef CONFIG_NUMA /* 3 seconds idle till flush */ |
fbc2edb05
|
749 |
__this_cpu_write(p->expire, 3); |
4037d4522
|
750 |
#endif |
2244b95a7
|
751 |
} |
fbc2edb05
|
752 |
} |
4037d4522
|
753 |
#ifdef CONFIG_NUMA |
3a321d2a3
|
754 755 756 757 758 759 760 761 762 763 764 |
for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) { int v; v = this_cpu_xchg(p->vm_numa_stat_diff[i], 0); if (v) { atomic_long_add(v, &zone->vm_numa_stat[i]); global_numa_diff[i] += v; __this_cpu_write(p->expire, 3); } } |
0eb77e988
|
765 766 767 768 769 770 771 772 773 774 |
if (do_pagesets) { cond_resched(); /* * Deal with draining the remote pageset of this * processor * * Check if there are pages remaining in this pageset * if not then there is nothing to expire. */ if (!__this_cpu_read(p->expire) || |
fbc2edb05
|
775 |
!__this_cpu_read(p->pcp.count)) |
0eb77e988
|
776 |
continue; |
4037d4522
|
777 |
|
0eb77e988
|
778 779 780 781 782 783 784 |
/* * We never drain zones local to this processor. */ if (zone_to_nid(zone) == numa_node_id()) { __this_cpu_write(p->expire, 0); continue; } |
4037d4522
|
785 |
|
0eb77e988
|
786 787 |
if (__this_cpu_dec_return(p->expire)) continue; |
4037d4522
|
788 |
|
0eb77e988
|
789 790 791 792 |
if (__this_cpu_read(p->pcp.count)) { drain_zone_pages(zone, this_cpu_ptr(&p->pcp)); changes++; } |
7cc36bbdd
|
793 |
} |
4037d4522
|
794 |
#endif |
2244b95a7
|
795 |
} |
75ef71840
|
796 797 798 799 800 801 802 803 804 805 806 807 808 809 |
for_each_online_pgdat(pgdat) { struct per_cpu_nodestat __percpu *p = pgdat->per_cpu_nodestats; for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) { int v; v = this_cpu_xchg(p->vm_node_stat_diff[i], 0); if (v) { atomic_long_add(v, &pgdat->vm_stat[i]); global_node_diff[i] += v; } } } |
3a321d2a3
|
810 811 812 813 |
#ifdef CONFIG_NUMA changes += fold_diff(global_zone_diff, global_numa_diff, global_node_diff); #else |
75ef71840
|
814 |
changes += fold_diff(global_zone_diff, global_node_diff); |
3a321d2a3
|
815 |
#endif |
7cc36bbdd
|
816 |
return changes; |
2244b95a7
|
817 |
} |
40f4b1ead
|
818 |
/* |
2bb921e52
|
819 820 821 822 823 824 |
* Fold the data for an offline cpu into the global array. * There cannot be any access by the offline cpu and therefore * synchronization is simplified. */ void cpu_vm_stats_fold(int cpu) { |
75ef71840
|
825 |
struct pglist_data *pgdat; |
2bb921e52
|
826 827 |
struct zone *zone; int i; |
75ef71840
|
828 |
int global_zone_diff[NR_VM_ZONE_STAT_ITEMS] = { 0, }; |
3a321d2a3
|
829 830 831 |
#ifdef CONFIG_NUMA int global_numa_diff[NR_VM_NUMA_STAT_ITEMS] = { 0, }; #endif |
75ef71840
|
832 |
int global_node_diff[NR_VM_NODE_STAT_ITEMS] = { 0, }; |
2bb921e52
|
833 834 835 836 837 838 839 840 841 842 843 844 845 |
for_each_populated_zone(zone) { struct per_cpu_pageset *p; p = per_cpu_ptr(zone->pageset, cpu); for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) if (p->vm_stat_diff[i]) { int v; v = p->vm_stat_diff[i]; p->vm_stat_diff[i] = 0; atomic_long_add(v, &zone->vm_stat[i]); |
75ef71840
|
846 |
global_zone_diff[i] += v; |
2bb921e52
|
847 |
} |
3a321d2a3
|
848 849 850 851 852 853 854 855 856 857 858 859 |
#ifdef CONFIG_NUMA for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) if (p->vm_numa_stat_diff[i]) { int v; v = p->vm_numa_stat_diff[i]; p->vm_numa_stat_diff[i] = 0; atomic_long_add(v, &zone->vm_numa_stat[i]); global_numa_diff[i] += v; } #endif |
2bb921e52
|
860 |
} |
75ef71840
|
861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 |
for_each_online_pgdat(pgdat) { struct per_cpu_nodestat *p; p = per_cpu_ptr(pgdat->per_cpu_nodestats, cpu); for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) if (p->vm_node_stat_diff[i]) { int v; v = p->vm_node_stat_diff[i]; p->vm_node_stat_diff[i] = 0; atomic_long_add(v, &pgdat->vm_stat[i]); global_node_diff[i] += v; } } |
3a321d2a3
|
876 877 878 |
#ifdef CONFIG_NUMA fold_diff(global_zone_diff, global_numa_diff, global_node_diff); #else |
75ef71840
|
879 |
fold_diff(global_zone_diff, global_node_diff); |
3a321d2a3
|
880 |
#endif |
2bb921e52
|
881 882 883 |
} /* |
40f4b1ead
|
884 885 886 |
* this is only called if !populated_zone(zone), which implies no other users of * pset->vm_stat_diff[] exsist. */ |
5a8838138
|
887 888 889 890 891 892 893 894 895 |
void drain_zonestat(struct zone *zone, struct per_cpu_pageset *pset) { int i; for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) if (pset->vm_stat_diff[i]) { int v = pset->vm_stat_diff[i]; pset->vm_stat_diff[i] = 0; atomic_long_add(v, &zone->vm_stat[i]); |
75ef71840
|
896 |
atomic_long_add(v, &vm_zone_stat[i]); |
5a8838138
|
897 |
} |
3a321d2a3
|
898 899 900 901 902 903 904 905 906 907 908 |
#ifdef CONFIG_NUMA for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) if (pset->vm_numa_stat_diff[i]) { int v = pset->vm_numa_stat_diff[i]; pset->vm_numa_stat_diff[i] = 0; atomic_long_add(v, &zone->vm_numa_stat[i]); atomic_long_add(v, &vm_numa_stat[i]); } #endif |
5a8838138
|
909 |
} |
2244b95a7
|
910 |
#endif |
ca889e6c4
|
911 |
#ifdef CONFIG_NUMA |
3a321d2a3
|
912 913 914 915 |
void __inc_numa_state(struct zone *zone, enum numa_stat_item item) { struct per_cpu_pageset __percpu *pcp = zone->pageset; |
1d90ca897
|
916 917 |
u16 __percpu *p = pcp->vm_numa_stat_diff + item; u16 v; |
3a321d2a3
|
918 919 |
v = __this_cpu_inc_return(*p); |
3a321d2a3
|
920 |
|
1d90ca897
|
921 922 923 |
if (unlikely(v > NUMA_STATS_THRESHOLD)) { zone_numa_state_add(v, zone, item); __this_cpu_write(*p, 0); |
3a321d2a3
|
924 925 |
} } |
ca889e6c4
|
926 |
/* |
75ef71840
|
927 928 929 |
* Determine the per node value of a stat item. This function * is called frequently in a NUMA machine, so try to be as * frugal as possible. |
c2d42c16a
|
930 |
*/ |
75ef71840
|
931 932 |
unsigned long sum_zone_node_page_state(int node, enum zone_stat_item item) |
c2d42c16a
|
933 934 |
{ struct zone *zones = NODE_DATA(node)->node_zones; |
e87d59f7a
|
935 936 |
int i; unsigned long count = 0; |
c2d42c16a
|
937 |
|
e87d59f7a
|
938 939 940 941 |
for (i = 0; i < MAX_NR_ZONES; i++) count += zone_page_state(zones + i, item); return count; |
c2d42c16a
|
942 |
} |
638032224
|
943 944 945 946 |
/* * Determine the per node value of a numa stat item. To avoid deviation, * the per cpu stat number in vm_numa_stat_diff[] is also included. */ |
3a321d2a3
|
947 948 949 950 951 952 953 954 |
unsigned long sum_zone_numa_state(int node, enum numa_stat_item item) { struct zone *zones = NODE_DATA(node)->node_zones; int i; unsigned long count = 0; for (i = 0; i < MAX_NR_ZONES; i++) |
638032224
|
955 |
count += zone_numa_state_snapshot(zones + i, item); |
3a321d2a3
|
956 957 958 |
return count; } |
75ef71840
|
959 960 961 |
/* * Determine the per node value of a stat item. */ |
ea426c2a7
|
962 963 |
unsigned long node_page_state_pages(struct pglist_data *pgdat, enum node_stat_item item) |
75ef71840
|
964 965 966 967 968 969 970 971 |
{ long x = atomic_long_read(&pgdat->vm_stat[item]); #ifdef CONFIG_SMP if (x < 0) x = 0; #endif return x; } |
ea426c2a7
|
972 973 974 975 976 977 978 979 |
unsigned long node_page_state(struct pglist_data *pgdat, enum node_stat_item item) { VM_WARN_ON_ONCE(vmstat_item_in_bytes(item)); return node_page_state_pages(pgdat, item); } |
ca889e6c4
|
980 |
#endif |
d7a5752c0
|
981 |
#ifdef CONFIG_COMPACTION |
36deb0be3
|
982 |
|
d7a5752c0
|
983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 |
struct contig_page_info { unsigned long free_pages; unsigned long free_blocks_total; unsigned long free_blocks_suitable; }; /* * Calculate the number of free pages in a zone, how many contiguous * pages are free and how many are large enough to satisfy an allocation of * the target size. Note that this function makes no attempt to estimate * how many suitable free blocks there *might* be if MOVABLE pages were * migrated. Calculating that is possible, but expensive and can be * figured out from userspace */ static void fill_contig_page_info(struct zone *zone, unsigned int suitable_order, struct contig_page_info *info) { unsigned int order; info->free_pages = 0; info->free_blocks_total = 0; info->free_blocks_suitable = 0; for (order = 0; order < MAX_ORDER; order++) { unsigned long blocks; /* Count number of free blocks */ blocks = zone->free_area[order].nr_free; info->free_blocks_total += blocks; /* Count free base pages */ info->free_pages += blocks << order; /* Count the suitable free blocks */ if (order >= suitable_order) info->free_blocks_suitable += blocks << (order - suitable_order); } } |
f1a5ab121
|
1023 1024 1025 1026 1027 1028 1029 1030 |
/* * A fragmentation index only makes sense if an allocation of a requested * size would fail. If that is true, the fragmentation index indicates * whether external fragmentation or a lack of memory was the problem. * The value can be used to determine if page reclaim or compaction * should be used */ |
56de7263f
|
1031 |
static int __fragmentation_index(unsigned int order, struct contig_page_info *info) |
f1a5ab121
|
1032 1033 |
{ unsigned long requested = 1UL << order; |
88d6ac40c
|
1034 1035 |
if (WARN_ON_ONCE(order >= MAX_ORDER)) return 0; |
f1a5ab121
|
1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 |
if (!info->free_blocks_total) return 0; /* Fragmentation index only makes sense when a request would fail */ if (info->free_blocks_suitable) return -1000; /* * Index is between 0 and 1 so return within 3 decimal places * * 0 => allocation would fail due to lack of memory * 1 => allocation would fail due to fragmentation */ return 1000 - div_u64( (1000+(div_u64(info->free_pages * 1000ULL, requested))), info->free_blocks_total); } |
56de7263f
|
1051 |
|
facdaa917
|
1052 1053 1054 1055 1056 |
/* * Calculates external fragmentation within a zone wrt the given order. * It is defined as the percentage of pages found in blocks of size * less than 1 << order. It returns values in range [0, 100]. */ |
d34c0a759
|
1057 |
unsigned int extfrag_for_order(struct zone *zone, unsigned int order) |
facdaa917
|
1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 |
{ struct contig_page_info info; fill_contig_page_info(zone, order, &info); if (info.free_pages == 0) return 0; return div_u64((info.free_pages - (info.free_blocks_suitable << order)) * 100, info.free_pages); } |
56de7263f
|
1069 1070 1071 1072 1073 1074 1075 1076 |
/* Same as __fragmentation index but allocs contig_page_info on stack */ int fragmentation_index(struct zone *zone, unsigned int order) { struct contig_page_info info; fill_contig_page_info(zone, order, &info); return __fragmentation_index(order, &info); } |
d7a5752c0
|
1077 |
#endif |
ebc5d83d0
|
1078 1079 |
#if defined(CONFIG_PROC_FS) || defined(CONFIG_SYSFS) || \ defined(CONFIG_NUMA) || defined(CONFIG_MEMCG) |
fa25c503d
|
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 |
#ifdef CONFIG_ZONE_DMA #define TEXT_FOR_DMA(xx) xx "_dma", #else #define TEXT_FOR_DMA(xx) #endif #ifdef CONFIG_ZONE_DMA32 #define TEXT_FOR_DMA32(xx) xx "_dma32", #else #define TEXT_FOR_DMA32(xx) #endif #ifdef CONFIG_HIGHMEM #define TEXT_FOR_HIGHMEM(xx) xx "_high", #else #define TEXT_FOR_HIGHMEM(xx) #endif #define TEXTS_FOR_ZONES(xx) TEXT_FOR_DMA(xx) TEXT_FOR_DMA32(xx) xx "_normal", \ TEXT_FOR_HIGHMEM(xx) xx "_movable", const char * const vmstat_text[] = { |
8d92890bd
|
1102 |
/* enum zone_stat_item counters */ |
fa25c503d
|
1103 |
"nr_free_pages", |
71c799f49
|
1104 1105 1106 1107 1108 |
"nr_zone_inactive_anon", "nr_zone_active_anon", "nr_zone_inactive_file", "nr_zone_active_file", "nr_zone_unevictable", |
5a1c84b40
|
1109 |
"nr_zone_write_pending", |
fa25c503d
|
1110 |
"nr_mlock", |
fa25c503d
|
1111 |
"nr_page_table_pages", |
fa25c503d
|
1112 |
"nr_bounce", |
91537fee0
|
1113 1114 1115 |
#if IS_ENABLED(CONFIG_ZSMALLOC) "nr_zspages", #endif |
3a321d2a3
|
1116 1117 1118 |
"nr_free_cma", /* enum numa_stat_item counters */ |
fa25c503d
|
1119 1120 1121 1122 1123 1124 1125 1126 |
#ifdef CONFIG_NUMA "numa_hit", "numa_miss", "numa_foreign", "numa_interleave", "numa_local", "numa_other", #endif |
09316c09d
|
1127 |
|
9d7ea9a29
|
1128 |
/* enum node_stat_item counters */ |
599d0c954
|
1129 1130 1131 1132 1133 |
"nr_inactive_anon", "nr_active_anon", "nr_inactive_file", "nr_active_file", "nr_unevictable", |
385386cff
|
1134 1135 |
"nr_slab_reclaimable", "nr_slab_unreclaimable", |
599d0c954
|
1136 1137 |
"nr_isolated_anon", "nr_isolated_file", |
68d48e6a2
|
1138 |
"workingset_nodes", |
170b04b7a
|
1139 1140 1141 1142 1143 1144 |
"workingset_refault_anon", "workingset_refault_file", "workingset_activate_anon", "workingset_activate_file", "workingset_restore_anon", "workingset_restore_file", |
1e6b10857
|
1145 |
"workingset_nodereclaim", |
50658e2e0
|
1146 1147 |
"nr_anon_pages", "nr_mapped", |
11fb99898
|
1148 1149 1150 1151 1152 1153 1154 |
"nr_file_pages", "nr_dirty", "nr_writeback", "nr_writeback_temp", "nr_shmem", "nr_shmem_hugepages", "nr_shmem_pmdmapped", |
60fbf0ab5
|
1155 1156 |
"nr_file_hugepages", "nr_file_pmdmapped", |
11fb99898
|
1157 |
"nr_anon_transparent_hugepages", |
c4a25635b
|
1158 1159 1160 1161 |
"nr_vmscan_write", "nr_vmscan_immediate_reclaim", "nr_dirtied", "nr_written", |
b29940c1a
|
1162 |
"nr_kernel_misc_reclaimable", |
1970dc6f5
|
1163 1164 |
"nr_foll_pin_acquired", "nr_foll_pin_released", |
991e76738
|
1165 1166 1167 1168 |
"nr_kernel_stack", #if IS_ENABLED(CONFIG_SHADOW_CALL_STACK) "nr_shadow_call_stack", #endif |
599d0c954
|
1169 |
|
09316c09d
|
1170 |
/* enum writeback_stat_item counters */ |
fa25c503d
|
1171 1172 |
"nr_dirty_threshold", "nr_dirty_background_threshold", |
ebc5d83d0
|
1173 |
#if defined(CONFIG_VM_EVENT_COUNTERS) || defined(CONFIG_MEMCG) |
09316c09d
|
1174 |
/* enum vm_event_item counters */ |
fa25c503d
|
1175 1176 1177 1178 1179 1180 |
"pgpgin", "pgpgout", "pswpin", "pswpout", TEXTS_FOR_ZONES("pgalloc") |
7cc30fcfd
|
1181 1182 |
TEXTS_FOR_ZONES("allocstall") TEXTS_FOR_ZONES("pgskip") |
fa25c503d
|
1183 1184 1185 1186 |
"pgfree", "pgactivate", "pgdeactivate", |
f7ad2a6cb
|
1187 |
"pglazyfree", |
fa25c503d
|
1188 1189 1190 |
"pgfault", "pgmajfault", |
854e9ed09
|
1191 |
"pglazyfreed", |
fa25c503d
|
1192 |
|
599d0c954
|
1193 1194 1195 1196 1197 |
"pgrefill", "pgsteal_kswapd", "pgsteal_direct", "pgscan_kswapd", "pgscan_direct", |
68243e76e
|
1198 |
"pgscan_direct_throttle", |
497a6c1b0
|
1199 1200 1201 1202 |
"pgscan_anon", "pgscan_file", "pgsteal_anon", "pgsteal_file", |
fa25c503d
|
1203 1204 1205 1206 1207 1208 |
#ifdef CONFIG_NUMA "zone_reclaim_failed", #endif "pginodesteal", "slabs_scanned", |
fa25c503d
|
1209 1210 1211 |
"kswapd_inodesteal", "kswapd_low_wmark_hit_quickly", "kswapd_high_wmark_hit_quickly", |
fa25c503d
|
1212 |
"pageoutrun", |
fa25c503d
|
1213 1214 |
"pgrotated", |
5509a5d27
|
1215 1216 |
"drop_pagecache", "drop_slab", |
8e675f7af
|
1217 |
"oom_kill", |
5509a5d27
|
1218 |
|
03c5a6e16
|
1219 1220 |
#ifdef CONFIG_NUMA_BALANCING "numa_pte_updates", |
72403b4a0
|
1221 |
"numa_huge_pte_updates", |
03c5a6e16
|
1222 1223 1224 1225 |
"numa_hint_faults", "numa_hint_faults_local", "numa_pages_migrated", #endif |
5647bc293
|
1226 1227 1228 |
#ifdef CONFIG_MIGRATION "pgmigrate_success", "pgmigrate_fail", |
1a5bae25e
|
1229 1230 1231 |
"thp_migration_success", "thp_migration_fail", "thp_migration_split", |
5647bc293
|
1232 |
#endif |
fa25c503d
|
1233 |
#ifdef CONFIG_COMPACTION |
397487db6
|
1234 1235 1236 |
"compact_migrate_scanned", "compact_free_scanned", "compact_isolated", |
fa25c503d
|
1237 1238 1239 |
"compact_stall", "compact_fail", "compact_success", |
698b1b306
|
1240 |
"compact_daemon_wake", |
7f354a548
|
1241 1242 |
"compact_daemon_migrate_scanned", "compact_daemon_free_scanned", |
fa25c503d
|
1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 |
#endif #ifdef CONFIG_HUGETLB_PAGE "htlb_buddy_alloc_success", "htlb_buddy_alloc_fail", #endif "unevictable_pgs_culled", "unevictable_pgs_scanned", "unevictable_pgs_rescued", "unevictable_pgs_mlocked", "unevictable_pgs_munlocked", "unevictable_pgs_cleared", "unevictable_pgs_stranded", |
fa25c503d
|
1256 1257 1258 1259 |
#ifdef CONFIG_TRANSPARENT_HUGEPAGE "thp_fault_alloc", "thp_fault_fallback", |
85b9f46e8
|
1260 |
"thp_fault_fallback_charge", |
fa25c503d
|
1261 1262 |
"thp_collapse_alloc", "thp_collapse_alloc_failed", |
95ecedcd6
|
1263 |
"thp_file_alloc", |
dcdf11ee1
|
1264 |
"thp_file_fallback", |
85b9f46e8
|
1265 |
"thp_file_fallback_charge", |
95ecedcd6
|
1266 |
"thp_file_mapped", |
122afea96
|
1267 1268 |
"thp_split_page", "thp_split_page_failed", |
f9719a03d
|
1269 |
"thp_deferred_split_page", |
122afea96
|
1270 |
"thp_split_pmd", |
ce9311cf9
|
1271 1272 1273 |
#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD "thp_split_pud", #endif |
d8a8e1f0d
|
1274 1275 |
"thp_zero_page_alloc", "thp_zero_page_alloc_failed", |
225311a46
|
1276 |
"thp_swpout", |
fe490cc0f
|
1277 |
"thp_swpout_fallback", |
fa25c503d
|
1278 |
#endif |
09316c09d
|
1279 1280 1281 1282 1283 1284 1285 |
#ifdef CONFIG_MEMORY_BALLOON "balloon_inflate", "balloon_deflate", #ifdef CONFIG_BALLOON_COMPACTION "balloon_migrate", #endif #endif /* CONFIG_MEMORY_BALLOON */ |
ec6599344
|
1286 |
#ifdef CONFIG_DEBUG_TLBFLUSH |
9824cf975
|
1287 1288 1289 1290 |
"nr_tlb_remote_flush", "nr_tlb_remote_flush_received", "nr_tlb_local_flush_all", "nr_tlb_local_flush_one", |
ec6599344
|
1291 |
#endif /* CONFIG_DEBUG_TLBFLUSH */ |
fa25c503d
|
1292 |
|
4f115147f
|
1293 1294 1295 1296 |
#ifdef CONFIG_DEBUG_VM_VMACACHE "vmacache_find_calls", "vmacache_find_hits", #endif |
cbc65df24
|
1297 1298 1299 1300 |
#ifdef CONFIG_SWAP "swap_ra", "swap_ra_hit", #endif |
ebc5d83d0
|
1301 |
#endif /* CONFIG_VM_EVENT_COUNTERS || CONFIG_MEMCG */ |
fa25c503d
|
1302 |
}; |
ebc5d83d0
|
1303 |
#endif /* CONFIG_PROC_FS || CONFIG_SYSFS || CONFIG_NUMA || CONFIG_MEMCG */ |
fa25c503d
|
1304 |
|
3c4868710
|
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 |
#if (defined(CONFIG_DEBUG_FS) && defined(CONFIG_COMPACTION)) || \ defined(CONFIG_PROC_FS) static void *frag_start(struct seq_file *m, loff_t *pos) { pg_data_t *pgdat; loff_t node = *pos; for (pgdat = first_online_pgdat(); pgdat && node; pgdat = next_online_pgdat(pgdat)) --node; return pgdat; } static void *frag_next(struct seq_file *m, void *arg, loff_t *pos) { pg_data_t *pgdat = (pg_data_t *)arg; (*pos)++; return next_online_pgdat(pgdat); } static void frag_stop(struct seq_file *m, void *arg) { } |
b2bd85981
|
1331 1332 1333 1334 |
/* * Walk zones in a node and print using a callback. * If @assert_populated is true, only use callback for zones that are populated. */ |
3c4868710
|
1335 |
static void walk_zones_in_node(struct seq_file *m, pg_data_t *pgdat, |
727c080f0
|
1336 |
bool assert_populated, bool nolock, |
3c4868710
|
1337 1338 1339 1340 1341 1342 1343 |
void (*print)(struct seq_file *m, pg_data_t *, struct zone *)) { struct zone *zone; struct zone *node_zones = pgdat->node_zones; unsigned long flags; for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; ++zone) { |
b2bd85981
|
1344 |
if (assert_populated && !populated_zone(zone)) |
3c4868710
|
1345 |
continue; |
727c080f0
|
1346 1347 |
if (!nolock) spin_lock_irqsave(&zone->lock, flags); |
3c4868710
|
1348 |
print(m, pgdat, zone); |
727c080f0
|
1349 1350 |
if (!nolock) spin_unlock_irqrestore(&zone->lock, flags); |
3c4868710
|
1351 1352 1353 |
} } #endif |
d7a5752c0
|
1354 |
#ifdef CONFIG_PROC_FS |
467c996c1
|
1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 |
static void frag_show_print(struct seq_file *m, pg_data_t *pgdat, struct zone *zone) { int order; seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name); for (order = 0; order < MAX_ORDER; ++order) seq_printf(m, "%6lu ", zone->free_area[order].nr_free); seq_putc(m, ' '); } /* * This walks the free areas for each zone. */ static int frag_show(struct seq_file *m, void *arg) { pg_data_t *pgdat = (pg_data_t *)arg; |
727c080f0
|
1373 |
walk_zones_in_node(m, pgdat, true, false, frag_show_print); |
467c996c1
|
1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 |
return 0; } static void pagetypeinfo_showfree_print(struct seq_file *m, pg_data_t *pgdat, struct zone *zone) { int order, mtype; for (mtype = 0; mtype < MIGRATE_TYPES; mtype++) { seq_printf(m, "Node %4d, zone %8s, type %12s ", pgdat->node_id, zone->name, migratetype_names[mtype]); for (order = 0; order < MAX_ORDER; ++order) { unsigned long freecount = 0; struct free_area *area; struct list_head *curr; |
93b3a6744
|
1391 |
bool overflow = false; |
467c996c1
|
1392 1393 |
area = &(zone->free_area[order]); |
93b3a6744
|
1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 |
list_for_each(curr, &area->free_list[mtype]) { /* * Cap the free_list iteration because it might * be really large and we are under a spinlock * so a long time spent here could trigger a * hard lockup detector. Anyway this is a * debugging tool so knowing there is a handful * of pages of this order should be more than * sufficient. */ if (++freecount >= 100000) { overflow = true; break; } } seq_printf(m, "%s%6lu ", overflow ? ">" : "", freecount); spin_unlock_irq(&zone->lock); cond_resched(); spin_lock_irq(&zone->lock); |
467c996c1
|
1413 |
} |
f6ac2354d
|
1414 1415 1416 |
seq_putc(m, ' '); } |
467c996c1
|
1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 |
} /* Print out the free pages at each order for each migatetype */ static int pagetypeinfo_showfree(struct seq_file *m, void *arg) { int order; pg_data_t *pgdat = (pg_data_t *)arg; /* Print header */ seq_printf(m, "%-43s ", "Free pages count per migrate type at order"); for (order = 0; order < MAX_ORDER; ++order) seq_printf(m, "%6d ", order); seq_putc(m, ' '); |
727c080f0
|
1431 |
walk_zones_in_node(m, pgdat, true, false, pagetypeinfo_showfree_print); |
467c996c1
|
1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 |
return 0; } static void pagetypeinfo_showblockcount_print(struct seq_file *m, pg_data_t *pgdat, struct zone *zone) { int mtype; unsigned long pfn; unsigned long start_pfn = zone->zone_start_pfn; |
108bcc96e
|
1442 |
unsigned long end_pfn = zone_end_pfn(zone); |
467c996c1
|
1443 1444 1445 1446 |
unsigned long count[MIGRATE_TYPES] = { 0, }; for (pfn = start_pfn; pfn < end_pfn; pfn += pageblock_nr_pages) { struct page *page; |
d336e94e4
|
1447 1448 |
page = pfn_to_online_page(pfn); if (!page) |
467c996c1
|
1449 |
continue; |
eb33575cf
|
1450 1451 |
/* Watch for unexpected holes punched in the memmap */ if (!memmap_valid_within(pfn, page, zone)) |
e80d6a248
|
1452 |
continue; |
eb33575cf
|
1453 |
|
a91c43c73
|
1454 1455 |
if (page_zone(page) != zone) continue; |
467c996c1
|
1456 |
mtype = get_pageblock_migratetype(page); |
e80d6a248
|
1457 1458 |
if (mtype < MIGRATE_TYPES) count[mtype]++; |
467c996c1
|
1459 1460 1461 1462 1463 1464 1465 1466 1467 |
} /* Print counts */ seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name); for (mtype = 0; mtype < MIGRATE_TYPES; mtype++) seq_printf(m, "%12lu ", count[mtype]); seq_putc(m, ' '); } |
f113e6412
|
1468 |
/* Print out the number of pageblocks for each migratetype */ |
467c996c1
|
1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 |
static int pagetypeinfo_showblockcount(struct seq_file *m, void *arg) { int mtype; pg_data_t *pgdat = (pg_data_t *)arg; seq_printf(m, " %-23s", "Number of blocks type "); for (mtype = 0; mtype < MIGRATE_TYPES; mtype++) seq_printf(m, "%12s ", migratetype_names[mtype]); seq_putc(m, ' '); |
727c080f0
|
1480 1481 |
walk_zones_in_node(m, pgdat, true, false, pagetypeinfo_showblockcount_print); |
467c996c1
|
1482 1483 1484 |
return 0; } |
48c96a368
|
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 |
/* * Print out the number of pageblocks for each migratetype that contain pages * of other types. This gives an indication of how well fallbacks are being * contained by rmqueue_fallback(). It requires information from PAGE_OWNER * to determine what is going on */ static void pagetypeinfo_showmixedcount(struct seq_file *m, pg_data_t *pgdat) { #ifdef CONFIG_PAGE_OWNER int mtype; |
7dd80b8af
|
1495 |
if (!static_branch_unlikely(&page_owner_inited)) |
48c96a368
|
1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 |
return; drain_all_pages(NULL); seq_printf(m, " %-23s", "Number of mixed blocks "); for (mtype = 0; mtype < MIGRATE_TYPES; mtype++) seq_printf(m, "%12s ", migratetype_names[mtype]); seq_putc(m, ' '); |
727c080f0
|
1506 1507 |
walk_zones_in_node(m, pgdat, true, true, pagetypeinfo_showmixedcount_print); |
48c96a368
|
1508 1509 |
#endif /* CONFIG_PAGE_OWNER */ } |
467c996c1
|
1510 1511 1512 1513 1514 1515 1516 |
/* * This prints out statistics in relation to grouping pages by mobility. * It is expensive to collect so do not constantly read the file. */ static int pagetypeinfo_show(struct seq_file *m, void *arg) { pg_data_t *pgdat = (pg_data_t *)arg; |
41b25a378
|
1517 |
/* check memoryless node */ |
a47b53c5f
|
1518 |
if (!node_state(pgdat->node_id, N_MEMORY)) |
41b25a378
|
1519 |
return 0; |
467c996c1
|
1520 1521 1522 1523 1524 1525 1526 1527 |
seq_printf(m, "Page block order: %d ", pageblock_order); seq_printf(m, "Pages per block: %lu ", pageblock_nr_pages); seq_putc(m, ' '); pagetypeinfo_showfree(m, pgdat); pagetypeinfo_showblockcount(m, pgdat); |
48c96a368
|
1528 |
pagetypeinfo_showmixedcount(m, pgdat); |
467c996c1
|
1529 |
|
f6ac2354d
|
1530 1531 |
return 0; } |
8f32f7e5a
|
1532 |
static const struct seq_operations fragmentation_op = { |
f6ac2354d
|
1533 1534 1535 1536 1537 |
.start = frag_start, .next = frag_next, .stop = frag_stop, .show = frag_show, }; |
74e2e8e8c
|
1538 |
static const struct seq_operations pagetypeinfo_op = { |
467c996c1
|
1539 1540 1541 1542 1543 |
.start = frag_start, .next = frag_next, .stop = frag_stop, .show = pagetypeinfo_show, }; |
e2ecc8a79
|
1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 |
static bool is_zone_first_populated(pg_data_t *pgdat, struct zone *zone) { int zid; for (zid = 0; zid < MAX_NR_ZONES; zid++) { struct zone *compare = &pgdat->node_zones[zid]; if (populated_zone(compare)) return zone == compare; } |
e2ecc8a79
|
1554 1555 |
return false; } |
467c996c1
|
1556 1557 |
static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat, struct zone *zone) |
f6ac2354d
|
1558 |
{ |
467c996c1
|
1559 1560 |
int i; seq_printf(m, "Node %d, zone %8s", pgdat->node_id, zone->name); |
e2ecc8a79
|
1561 1562 1563 1564 |
if (is_zone_first_populated(pgdat, zone)) { seq_printf(m, " per-node stats"); for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) { |
9d7ea9a29
|
1565 1566 |
seq_printf(m, " %-12s %lu", node_stat_name(i), |
ea426c2a7
|
1567 |
node_page_state_pages(pgdat, i)); |
e2ecc8a79
|
1568 1569 |
} } |
467c996c1
|
1570 1571 1572 1573 1574 1575 1576 1577 1578 |
seq_printf(m, " pages free %lu" " min %lu" " low %lu" " high %lu" |
467c996c1
|
1579 1580 |
" spanned %lu" |
9feedc9d8
|
1581 1582 1583 1584 |
" present %lu" " managed %lu", |
88f5acf88
|
1585 |
zone_page_state(zone, NR_FREE_PAGES), |
418589663
|
1586 1587 1588 |
min_wmark_pages(zone), low_wmark_pages(zone), high_wmark_pages(zone), |
467c996c1
|
1589 |
zone->spanned_pages, |
9feedc9d8
|
1590 |
zone->present_pages, |
9705bea5f
|
1591 |
zone_managed_pages(zone)); |
467c996c1
|
1592 |
|
467c996c1
|
1593 |
seq_printf(m, |
3484b2de9
|
1594 1595 |
" protection: (%ld", |
467c996c1
|
1596 1597 |
zone->lowmem_reserve[0]); for (i = 1; i < ARRAY_SIZE(zone->lowmem_reserve); i++) |
3484b2de9
|
1598 |
seq_printf(m, ", %ld", zone->lowmem_reserve[i]); |
7dfb8bf3b
|
1599 |
seq_putc(m, ')'); |
a8a4b7aea
|
1600 1601 1602 1603 1604 1605 |
/* If unpopulated, no other information is useful */ if (!populated_zone(zone)) { seq_putc(m, ' '); return; } |
7dfb8bf3b
|
1606 |
for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) |
9d7ea9a29
|
1607 1608 1609 |
seq_printf(m, " %-12s %lu", zone_stat_name(i), zone_page_state(zone, i)); |
7dfb8bf3b
|
1610 |
|
3a321d2a3
|
1611 1612 |
#ifdef CONFIG_NUMA for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) |
9d7ea9a29
|
1613 1614 1615 |
seq_printf(m, " %-12s %lu", numa_stat_name(i), zone_numa_state_snapshot(zone, i)); |
3a321d2a3
|
1616 |
#endif |
7dfb8bf3b
|
1617 1618 |
seq_printf(m, " pagesets"); |
467c996c1
|
1619 1620 |
for_each_online_cpu(i) { struct per_cpu_pageset *pageset; |
467c996c1
|
1621 |
|
99dcc3e5a
|
1622 |
pageset = per_cpu_ptr(zone->pageset, i); |
3dfa5721f
|
1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 |
seq_printf(m, " cpu: %i" " count: %i" " high: %i" " batch: %i", i, pageset->pcp.count, pageset->pcp.high, pageset->pcp.batch); |
df9ecaba3
|
1636 |
#ifdef CONFIG_SMP |
467c996c1
|
1637 1638 1639 |
seq_printf(m, " vm stats threshold: %d", pageset->stat_threshold); |
df9ecaba3
|
1640 |
#endif |
f6ac2354d
|
1641 |
} |
467c996c1
|
1642 |
seq_printf(m, |
599d0c954
|
1643 1644 |
" node_unreclaimable: %u" |
3a50d14d0
|
1645 1646 |
" start_pfn: %lu", |
c73322d09
|
1647 |
pgdat->kswapd_failures >= MAX_RECLAIM_RETRIES, |
3a50d14d0
|
1648 |
zone->zone_start_pfn); |
467c996c1
|
1649 1650 1651 1652 1653 |
seq_putc(m, ' '); } /* |
b2bd85981
|
1654 1655 1656 1657 |
* Output information about zones in @pgdat. All zones are printed regardless * of whether they are populated or not: lowmem_reserve_ratio operates on the * set of all zones and userspace would not be aware of such zones if they are * suppressed here (zoneinfo displays the effect of lowmem_reserve_ratio). |
467c996c1
|
1658 1659 1660 1661 |
*/ static int zoneinfo_show(struct seq_file *m, void *arg) { pg_data_t *pgdat = (pg_data_t *)arg; |
727c080f0
|
1662 |
walk_zones_in_node(m, pgdat, false, false, zoneinfo_show_print); |
f6ac2354d
|
1663 1664 |
return 0; } |
5c9fe6281
|
1665 |
static const struct seq_operations zoneinfo_op = { |
f6ac2354d
|
1666 1667 1668 1669 1670 1671 |
.start = frag_start, /* iterate over all zones. The same as in * fragmentation. */ .next = frag_next, .stop = frag_stop, .show = zoneinfo_show, }; |
9d7ea9a29
|
1672 1673 1674 1675 1676 1677 |
#define NR_VMSTAT_ITEMS (NR_VM_ZONE_STAT_ITEMS + \ NR_VM_NUMA_STAT_ITEMS + \ NR_VM_NODE_STAT_ITEMS + \ NR_VM_WRITEBACK_STAT_ITEMS + \ (IS_ENABLED(CONFIG_VM_EVENT_COUNTERS) ? \ NR_VM_EVENT_ITEMS : 0)) |
79da826ae
|
1678 |
|
f6ac2354d
|
1679 1680 |
static void *vmstat_start(struct seq_file *m, loff_t *pos) { |
2244b95a7
|
1681 |
unsigned long *v; |
9d7ea9a29
|
1682 |
int i; |
f6ac2354d
|
1683 |
|
9d7ea9a29
|
1684 |
if (*pos >= NR_VMSTAT_ITEMS) |
f6ac2354d
|
1685 |
return NULL; |
79da826ae
|
1686 |
|
9d7ea9a29
|
1687 1688 |
BUILD_BUG_ON(ARRAY_SIZE(vmstat_text) < NR_VMSTAT_ITEMS); v = kmalloc_array(NR_VMSTAT_ITEMS, sizeof(unsigned long), GFP_KERNEL); |
2244b95a7
|
1689 1690 |
m->private = v; if (!v) |
f6ac2354d
|
1691 |
return ERR_PTR(-ENOMEM); |
2244b95a7
|
1692 |
for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) |
c41f012ad
|
1693 |
v[i] = global_zone_page_state(i); |
79da826ae
|
1694 |
v += NR_VM_ZONE_STAT_ITEMS; |
3a321d2a3
|
1695 1696 1697 1698 1699 |
#ifdef CONFIG_NUMA for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) v[i] = global_numa_state(i); v += NR_VM_NUMA_STAT_ITEMS; #endif |
75ef71840
|
1700 |
for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) |
ea426c2a7
|
1701 |
v[i] = global_node_page_state_pages(i); |
75ef71840
|
1702 |
v += NR_VM_NODE_STAT_ITEMS; |
79da826ae
|
1703 1704 1705 |
global_dirty_limits(v + NR_DIRTY_BG_THRESHOLD, v + NR_DIRTY_THRESHOLD); v += NR_VM_WRITEBACK_STAT_ITEMS; |
f8891e5e1
|
1706 |
#ifdef CONFIG_VM_EVENT_COUNTERS |
79da826ae
|
1707 1708 1709 |
all_vm_events(v); v[PGPGIN] /= 2; /* sectors -> kbytes */ v[PGPGOUT] /= 2; |
f8891e5e1
|
1710 |
#endif |
ff8b16d7e
|
1711 |
return (unsigned long *)m->private + *pos; |
f6ac2354d
|
1712 1713 1714 1715 1716 |
} static void *vmstat_next(struct seq_file *m, void *arg, loff_t *pos) { (*pos)++; |
9d7ea9a29
|
1717 |
if (*pos >= NR_VMSTAT_ITEMS) |
f6ac2354d
|
1718 1719 1720 1721 1722 1723 1724 1725 |
return NULL; return (unsigned long *)m->private + *pos; } static int vmstat_show(struct seq_file *m, void *arg) { unsigned long *l = arg; unsigned long off = l - (unsigned long *)m->private; |
68ba0326b
|
1726 1727 |
seq_puts(m, vmstat_text[off]); |
75ba1d07f
|
1728 |
seq_put_decimal_ull(m, " ", *l); |
68ba0326b
|
1729 1730 |
seq_putc(m, ' '); |
8d92890bd
|
1731 1732 1733 1734 1735 1736 1737 1738 1739 |
if (off == NR_VMSTAT_ITEMS - 1) { /* * We've come to the end - add any deprecated counters to avoid * breaking userspace which might depend on them being present. */ seq_puts(m, "nr_unstable 0 "); } |
f6ac2354d
|
1740 1741 1742 1743 1744 1745 1746 1747 |
return 0; } static void vmstat_stop(struct seq_file *m, void *arg) { kfree(m->private); m->private = NULL; } |
b6aa44ab6
|
1748 |
static const struct seq_operations vmstat_op = { |
f6ac2354d
|
1749 1750 1751 1752 1753 |
.start = vmstat_start, .next = vmstat_next, .stop = vmstat_stop, .show = vmstat_show, }; |
f6ac2354d
|
1754 |
#endif /* CONFIG_PROC_FS */ |
df9ecaba3
|
1755 |
#ifdef CONFIG_SMP |
d1187ed21
|
1756 |
static DEFINE_PER_CPU(struct delayed_work, vmstat_work); |
77461ab33
|
1757 |
int sysctl_stat_interval __read_mostly = HZ; |
d1187ed21
|
1758 |
|
52b6f46bc
|
1759 1760 1761 1762 1763 1764 1765 |
#ifdef CONFIG_PROC_FS static void refresh_vm_stats(struct work_struct *work) { refresh_cpu_vm_stats(true); } int vmstat_refresh(struct ctl_table *table, int write, |
32927393d
|
1766 |
void *buffer, size_t *lenp, loff_t *ppos) |
52b6f46bc
|
1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 |
{ long val; int err; int i; /* * The regular update, every sysctl_stat_interval, may come later * than expected: leaving a significant amount in per_cpu buckets. * This is particularly misleading when checking a quantity of HUGE * pages, immediately after running a test. /proc/sys/vm/stat_refresh, * which can equally be echo'ed to or cat'ted from (by root), * can be used to update the stats just before reading them. * |
c41f012ad
|
1780 |
* Oh, and since global_zone_page_state() etc. are so careful to hide |
52b6f46bc
|
1781 1782 1783 1784 1785 1786 1787 |
* transiently negative values, report an error here if any of * the stats is negative, so we know to go looking for imbalance. */ err = schedule_on_each_cpu(refresh_vm_stats); if (err) return err; for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) { |
75ef71840
|
1788 |
val = atomic_long_read(&vm_zone_stat[i]); |
52b6f46bc
|
1789 |
if (val < 0) { |
c822f6223
|
1790 1791 |
pr_warn("%s: %s %ld ", |
9d7ea9a29
|
1792 |
__func__, zone_stat_name(i), val); |
c822f6223
|
1793 |
err = -EINVAL; |
52b6f46bc
|
1794 1795 |
} } |
3a321d2a3
|
1796 1797 1798 1799 1800 1801 |
#ifdef CONFIG_NUMA for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) { val = atomic_long_read(&vm_numa_stat[i]); if (val < 0) { pr_warn("%s: %s %ld ", |
9d7ea9a29
|
1802 |
__func__, numa_stat_name(i), val); |
3a321d2a3
|
1803 1804 1805 1806 |
err = -EINVAL; } } #endif |
52b6f46bc
|
1807 1808 1809 1810 1811 1812 1813 1814 1815 |
if (err) return err; if (write) *ppos += *lenp; else *lenp = 0; return 0; } #endif /* CONFIG_PROC_FS */ |
d1187ed21
|
1816 1817 |
static void vmstat_update(struct work_struct *w) { |
0eb77e988
|
1818 |
if (refresh_cpu_vm_stats(true)) { |
7cc36bbdd
|
1819 1820 1821 1822 1823 |
/* * Counters were updated so we expect more updates * to occur in the future. Keep on running the * update worker thread. */ |
ce612879d
|
1824 |
queue_delayed_work_on(smp_processor_id(), mm_percpu_wq, |
f01f17d37
|
1825 1826 |
this_cpu_ptr(&vmstat_work), round_jiffies_relative(sysctl_stat_interval)); |
7cc36bbdd
|
1827 1828 1829 1830 |
} } /* |
0eb77e988
|
1831 1832 1833 1834 |
* Switch off vmstat processing and then fold all the remaining differentials * until the diffs stay at zero. The function is used by NOHZ and can only be * invoked when tick processing is not active. */ |
0eb77e988
|
1835 |
/* |
7cc36bbdd
|
1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 |
* Check if the diffs for a certain cpu indicate that * an update is needed. */ static bool need_update(int cpu) { struct zone *zone; for_each_populated_zone(zone) { struct per_cpu_pageset *p = per_cpu_ptr(zone->pageset, cpu); BUILD_BUG_ON(sizeof(p->vm_stat_diff[0]) != 1); |
3a321d2a3
|
1847 |
#ifdef CONFIG_NUMA |
1d90ca897
|
1848 |
BUILD_BUG_ON(sizeof(p->vm_numa_stat_diff[0]) != 2); |
3a321d2a3
|
1849 |
#endif |
638032224
|
1850 |
|
7cc36bbdd
|
1851 1852 |
/* * The fast way of checking if there are any vmstat diffs. |
7cc36bbdd
|
1853 |
*/ |
13c9aaf7f
|
1854 1855 |
if (memchr_inv(p->vm_stat_diff, 0, NR_VM_ZONE_STAT_ITEMS * sizeof(p->vm_stat_diff[0]))) |
7cc36bbdd
|
1856 |
return true; |
3a321d2a3
|
1857 |
#ifdef CONFIG_NUMA |
13c9aaf7f
|
1858 1859 |
if (memchr_inv(p->vm_numa_stat_diff, 0, NR_VM_NUMA_STAT_ITEMS * sizeof(p->vm_numa_stat_diff[0]))) |
3a321d2a3
|
1860 1861 |
return true; #endif |
7cc36bbdd
|
1862 1863 1864 |
} return false; } |
7b8da4c7f
|
1865 1866 1867 1868 1869 |
/* * Switch off vmstat processing and then fold all the remaining differentials * until the diffs stay at zero. The function is used by NOHZ and can only be * invoked when tick processing is not active. */ |
f01f17d37
|
1870 1871 1872 1873 |
void quiet_vmstat(void) { if (system_state != SYSTEM_RUNNING) return; |
7b8da4c7f
|
1874 |
if (!delayed_work_pending(this_cpu_ptr(&vmstat_work))) |
f01f17d37
|
1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 |
return; if (!need_update(smp_processor_id())) return; /* * Just refresh counters and do not care about the pending delayed * vmstat_update. It doesn't fire that often to matter and canceling * it would be too expensive from this path. * vmstat_shepherd will take care about that for us. */ refresh_cpu_vm_stats(false); } |
7cc36bbdd
|
1888 1889 1890 1891 1892 1893 1894 |
/* * Shepherd worker thread that checks the * differentials of processors that have their worker * threads for vm statistics updates disabled because of * inactivity. */ static void vmstat_shepherd(struct work_struct *w); |
0eb77e988
|
1895 |
static DECLARE_DEFERRABLE_WORK(shepherd, vmstat_shepherd); |
7cc36bbdd
|
1896 1897 1898 1899 1900 1901 1902 |
static void vmstat_shepherd(struct work_struct *w) { int cpu; get_online_cpus(); /* Check processors whose vmstat worker threads have been disabled */ |
7b8da4c7f
|
1903 |
for_each_online_cpu(cpu) { |
f01f17d37
|
1904 |
struct delayed_work *dw = &per_cpu(vmstat_work, cpu); |
7cc36bbdd
|
1905 |
|
7b8da4c7f
|
1906 |
if (!delayed_work_pending(dw) && need_update(cpu)) |
ce612879d
|
1907 |
queue_delayed_work_on(cpu, mm_percpu_wq, dw, 0); |
f01f17d37
|
1908 |
} |
7cc36bbdd
|
1909 1910 1911 |
put_online_cpus(); schedule_delayed_work(&shepherd, |
98f4ebb29
|
1912 |
round_jiffies_relative(sysctl_stat_interval)); |
d1187ed21
|
1913 |
} |
7cc36bbdd
|
1914 |
static void __init start_shepherd_timer(void) |
d1187ed21
|
1915 |
{ |
7cc36bbdd
|
1916 1917 1918 |
int cpu; for_each_possible_cpu(cpu) |
ccde8bd40
|
1919 |
INIT_DEFERRABLE_WORK(per_cpu_ptr(&vmstat_work, cpu), |
7cc36bbdd
|
1920 |
vmstat_update); |
7cc36bbdd
|
1921 1922 |
schedule_delayed_work(&shepherd, round_jiffies_relative(sysctl_stat_interval)); |
d1187ed21
|
1923 |
} |
03e86dba5
|
1924 1925 |
static void __init init_cpu_node_state(void) { |
4c501327b
|
1926 |
int node; |
03e86dba5
|
1927 |
|
4c501327b
|
1928 1929 1930 1931 |
for_each_online_node(node) { if (cpumask_weight(cpumask_of_node(node)) > 0) node_set_state(node, N_CPU); } |
03e86dba5
|
1932 |
} |
5438da977
|
1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 |
static int vmstat_cpu_online(unsigned int cpu) { refresh_zone_stat_thresholds(); node_set_state(cpu_to_node(cpu), N_CPU); return 0; } static int vmstat_cpu_down_prep(unsigned int cpu) { cancel_delayed_work_sync(&per_cpu(vmstat_work, cpu)); return 0; } static int vmstat_cpu_dead(unsigned int cpu) |
807a1bd2b
|
1947 |
{ |
4c501327b
|
1948 |
const struct cpumask *node_cpus; |
5438da977
|
1949 |
int node; |
807a1bd2b
|
1950 |
|
5438da977
|
1951 1952 1953 |
node = cpu_to_node(cpu); refresh_zone_stat_thresholds(); |
4c501327b
|
1954 1955 |
node_cpus = cpumask_of_node(node); if (cpumask_weight(node_cpus) > 0) |
5438da977
|
1956 |
return 0; |
807a1bd2b
|
1957 1958 |
node_clear_state(node, N_CPU); |
5438da977
|
1959 |
return 0; |
807a1bd2b
|
1960 |
} |
8f32f7e5a
|
1961 |
#endif |
df9ecaba3
|
1962 |
|
ce612879d
|
1963 |
struct workqueue_struct *mm_percpu_wq; |
597b7305d
|
1964 |
void __init init_mm_internals(void) |
df9ecaba3
|
1965 |
{ |
ce612879d
|
1966 |
int ret __maybe_unused; |
5438da977
|
1967 |
|
80d136e13
|
1968 |
mm_percpu_wq = alloc_workqueue("mm_percpu_wq", WQ_MEM_RECLAIM, 0); |
ce612879d
|
1969 1970 |
#ifdef CONFIG_SMP |
5438da977
|
1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 |
ret = cpuhp_setup_state_nocalls(CPUHP_MM_VMSTAT_DEAD, "mm/vmstat:dead", NULL, vmstat_cpu_dead); if (ret < 0) pr_err("vmstat: failed to register 'dead' hotplug state "); ret = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, "mm/vmstat:online", vmstat_cpu_online, vmstat_cpu_down_prep); if (ret < 0) pr_err("vmstat: failed to register 'online' hotplug state "); get_online_cpus(); |
03e86dba5
|
1985 |
init_cpu_node_state(); |
5438da977
|
1986 |
put_online_cpus(); |
d1187ed21
|
1987 |
|
7cc36bbdd
|
1988 |
start_shepherd_timer(); |
8f32f7e5a
|
1989 1990 |
#endif #ifdef CONFIG_PROC_FS |
fddda2b7b
|
1991 |
proc_create_seq("buddyinfo", 0444, NULL, &fragmentation_op); |
abaed0112
|
1992 |
proc_create_seq("pagetypeinfo", 0400, NULL, &pagetypeinfo_op); |
fddda2b7b
|
1993 1994 |
proc_create_seq("vmstat", 0444, NULL, &vmstat_op); proc_create_seq("zoneinfo", 0444, NULL, &zoneinfo_op); |
8f32f7e5a
|
1995 |
#endif |
df9ecaba3
|
1996 |
} |
d7a5752c0
|
1997 1998 |
#if defined(CONFIG_DEBUG_FS) && defined(CONFIG_COMPACTION) |
d7a5752c0
|
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 |
/* * Return an index indicating how much of the available free memory is * unusable for an allocation of the requested size. */ static int unusable_free_index(unsigned int order, struct contig_page_info *info) { /* No free memory is interpreted as all free memory is unusable */ if (info->free_pages == 0) return 1000; /* * Index should be a value between 0 and 1. Return a value to 3 * decimal places. * * 0 => no fragmentation * 1 => high fragmentation */ return div_u64((info->free_pages - (info->free_blocks_suitable << order)) * 1000ULL, info->free_pages); } static void unusable_show_print(struct seq_file *m, pg_data_t *pgdat, struct zone *zone) { unsigned int order; int index; struct contig_page_info info; seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name); for (order = 0; order < MAX_ORDER; ++order) { fill_contig_page_info(zone, order, &info); index = unusable_free_index(order, &info); seq_printf(m, "%d.%03d ", index / 1000, index % 1000); } seq_putc(m, ' '); } /* * Display unusable free space index * * The unusable free space index measures how much of the available free * memory cannot be used to satisfy an allocation of a given size and is a * value between 0 and 1. The higher the value, the more of free memory is * unusable and by implication, the worse the external fragmentation is. This * can be expressed as a percentage by multiplying by 100. */ static int unusable_show(struct seq_file *m, void *arg) { pg_data_t *pgdat = (pg_data_t *)arg; /* check memoryless node */ |
a47b53c5f
|
2056 |
if (!node_state(pgdat->node_id, N_MEMORY)) |
d7a5752c0
|
2057 |
return 0; |
727c080f0
|
2058 |
walk_zones_in_node(m, pgdat, true, false, unusable_show_print); |
d7a5752c0
|
2059 2060 2061 |
return 0; } |
01a995600
|
2062 |
static const struct seq_operations unusable_sops = { |
d7a5752c0
|
2063 2064 2065 2066 2067 |
.start = frag_start, .next = frag_next, .stop = frag_stop, .show = unusable_show, }; |
01a995600
|
2068 |
DEFINE_SEQ_ATTRIBUTE(unusable); |
d7a5752c0
|
2069 |
|
f1a5ab121
|
2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 |
static void extfrag_show_print(struct seq_file *m, pg_data_t *pgdat, struct zone *zone) { unsigned int order; int index; /* Alloc on stack as interrupts are disabled for zone walk */ struct contig_page_info info; seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name); for (order = 0; order < MAX_ORDER; ++order) { fill_contig_page_info(zone, order, &info); |
56de7263f
|
2084 |
index = __fragmentation_index(order, &info); |
f1a5ab121
|
2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 |
seq_printf(m, "%d.%03d ", index / 1000, index % 1000); } seq_putc(m, ' '); } /* * Display fragmentation index for orders that allocations would fail for */ static int extfrag_show(struct seq_file *m, void *arg) { pg_data_t *pgdat = (pg_data_t *)arg; |
727c080f0
|
2098 |
walk_zones_in_node(m, pgdat, true, false, extfrag_show_print); |
f1a5ab121
|
2099 2100 2101 |
return 0; } |
01a995600
|
2102 |
static const struct seq_operations extfrag_sops = { |
f1a5ab121
|
2103 2104 2105 2106 2107 |
.start = frag_start, .next = frag_next, .stop = frag_stop, .show = extfrag_show, }; |
01a995600
|
2108 |
DEFINE_SEQ_ATTRIBUTE(extfrag); |
f1a5ab121
|
2109 |
|
d7a5752c0
|
2110 2111 |
static int __init extfrag_debug_init(void) { |
bde8bd8a1
|
2112 |
struct dentry *extfrag_debug_root; |
d7a5752c0
|
2113 |
extfrag_debug_root = debugfs_create_dir("extfrag", NULL); |
d7a5752c0
|
2114 |
|
d9f7979c9
|
2115 |
debugfs_create_file("unusable_index", 0444, extfrag_debug_root, NULL, |
01a995600
|
2116 |
&unusable_fops); |
d7a5752c0
|
2117 |
|
d9f7979c9
|
2118 |
debugfs_create_file("extfrag_index", 0444, extfrag_debug_root, NULL, |
01a995600
|
2119 |
&extfrag_fops); |
f1a5ab121
|
2120 |
|
d7a5752c0
|
2121 2122 2123 2124 2125 |
return 0; } module_init(extfrag_debug_init); #endif |