#include <linux/memory_hotplug.h>
#include <linux/nodemask.h>
#include <linux/vmalloc.h>
+#include <linux/mempolicy.h>
#include <asm/tlbflush.h>
#include "internal.h"
unsigned long totalram_pages __read_mostly;
unsigned long totalhigh_pages __read_mostly;
long nr_swap_pages;
+int percpu_pagelist_fraction;
static void fastcall free_hot_cold_page(struct page *page, int cold);
+static void __free_pages_ok(struct page *page, unsigned int order);
/*
* results with 256, 32 in the lowmem_reserve sysctl:
static void bad_page(struct page *page)
{
printk(KERN_EMERG "Bad page state in process '%s'\n"
- "page:%p flags:0x%0*lx mapping:%p mapcount:%d count:%d\n"
- "Trying to fix it up, but a reboot is needed\n"
- "Backtrace:\n",
+ KERN_EMERG "page:%p flags:0x%0*lx mapping:%p mapcount:%d count:%d\n"
+ KERN_EMERG "Trying to fix it up, but a reboot is needed\n"
+ KERN_EMERG "Backtrace:\n",
current->comm, page, (int)(2*sizeof(unsigned long)),
(unsigned long)page->flags, page->mapping,
page_mapcount(page), page_count(page));
* All pages have PG_compound set. All pages have their ->private pointing at
* the head page (even the head page has this).
*
- * The first tail page's ->mapping, if non-zero, holds the address of the
- * compound page's put_page() function.
- *
- * The order of the allocation is stored in the first tail page's ->index
- * This is only for debug at present. This usage means that zero-order pages
- * may not be compound.
+ * 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.
*/
+
+static void free_compound_page(struct page *page)
+{
+ __free_pages_ok(page, (unsigned long)page[1].lru.prev);
+}
+
static void prep_compound_page(struct page *page, unsigned long order)
{
int i;
int nr_pages = 1 << order;
- page[1].mapping = NULL;
- page[1].index = order;
+ page[1].lru.next = (void *)free_compound_page; /* set dtor */
+ page[1].lru.prev = (void *)order;
for (i = 0; i < nr_pages; i++) {
struct page *p = page + i;
int i;
int nr_pages = 1 << order;
- if (unlikely(page[1].index != order))
+ if (unlikely((unsigned long)page[1].lru.prev != order))
bad_page(page);
for (i = 0; i < nr_pages; i++) {
* -- wli
*/
-static inline void __free_pages_bulk (struct page *page,
+static inline void __free_one_page(struct page *page,
struct zone *zone, unsigned int order)
{
unsigned long page_idx;
* And clear the zone's pages_scanned counter, to hold off the "all pages are
* pinned" detection logic.
*/
-static int
-free_pages_bulk(struct zone *zone, int count,
- struct list_head *list, unsigned int order)
+static void free_pages_bulk(struct zone *zone, int count,
+ struct list_head *list, int order)
{
- struct page *page = NULL;
- int ret = 0;
-
spin_lock(&zone->lock);
zone->all_unreclaimable = 0;
zone->pages_scanned = 0;
- while (!list_empty(list) && count--) {
+ while (count--) {
+ struct page *page;
+
+ BUG_ON(list_empty(list));
page = list_entry(list->prev, struct page, lru);
- /* have to delete it as __free_pages_bulk list manipulates */
+ /* have to delete it as __free_one_page list manipulates */
list_del(&page->lru);
- __free_pages_bulk(page, zone, order);
- ret++;
+ __free_one_page(page, zone, order);
}
spin_unlock(&zone->lock);
- return ret;
}
-void __free_pages_ok(struct page *page, unsigned int order)
+static void free_one_page(struct zone *zone, struct page *page, int order)
{
- unsigned long flags;
LIST_HEAD(list);
+ list_add(&page->lru, &list);
+ free_pages_bulk(zone, 1, &list, order);
+}
+
+static void __free_pages_ok(struct page *page, unsigned int order)
+{
+ unsigned long flags;
int i;
int reserved = 0;
arch_free_page(page, order);
+ if (!PageHighMem(page))
+ mutex_debug_check_no_locks_freed(page_address(page),
+ PAGE_SIZE<<order);
#ifndef CONFIG_MMU
- if (order > 0)
- for (i = 1 ; i < (1 << order) ; ++i)
- __put_page(page + i);
+ for (i = 1 ; i < (1 << order) ; ++i)
+ __put_page(page + i);
#endif
for (i = 0 ; i < (1 << order) ; ++i)
if (reserved)
return;
- list_add(&page->lru, &list);
- mod_page_state(pgfree, 1 << order);
- kernel_map_pages(page, 1<<order, 0);
+ kernel_map_pages(page, 1 << order, 0);
local_irq_save(flags);
- free_pages_bulk(page_zone(page), 1, &list, order);
+ __mod_page_state(pgfree, 1 << order);
+ free_one_page(page_zone(page), page, order);
local_irq_restore(flags);
}
}
#ifdef CONFIG_NUMA
-/* Called from the slab reaper to drain remote pagesets */
-void drain_remote_pages(void)
+/*
+ * Called from the slab reaper to drain pagesets on a particular node that
+ * belong to the currently executing processor.
+ */
+void drain_node_pages(int nodeid)
{
- struct zone *zone;
- int i;
+ int i, z;
unsigned long flags;
local_irq_save(flags);
- for_each_zone(zone) {
+ for (z = 0; z < MAX_NR_ZONES; z++) {
+ struct zone *zone = NODE_DATA(nodeid)->node_zones + z;
struct per_cpu_pageset *pset;
- /* Do not drain local pagesets */
- if (zone->zone_pgdat->node_id == numa_node_id())
- continue;
-
- pset = zone->pageset[smp_processor_id()];
+ pset = zone_pcp(zone, smp_processor_id());
for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) {
struct per_cpu_pages *pcp;
pcp = &pset->pcp[i];
- if (pcp->count)
- pcp->count -= free_pages_bulk(zone, pcp->count,
- &pcp->list, 0);
+ free_pages_bulk(zone, pcp->count, &pcp->list, 0);
+ pcp->count = 0;
}
}
local_irq_restore(flags);
pcp = &pset->pcp[i];
local_irq_save(flags);
- pcp->count -= free_pages_bulk(zone, pcp->count,
- &pcp->list, 0);
+ free_pages_bulk(zone, pcp->count, &pcp->list, 0);
+ pcp->count = 0;
local_irq_restore(flags);
}
}
}
#endif /* CONFIG_PM */
-static void zone_statistics(struct zonelist *zonelist, struct zone *z)
+static void zone_statistics(struct zonelist *zonelist, struct zone *z, int cpu)
{
#ifdef CONFIG_NUMA
- unsigned long flags;
- int cpu;
pg_data_t *pg = z->zone_pgdat;
pg_data_t *orig = zonelist->zones[0]->zone_pgdat;
struct per_cpu_pageset *p;
- local_irq_save(flags);
- cpu = smp_processor_id();
- p = zone_pcp(z,cpu);
+ p = zone_pcp(z, cpu);
if (pg == orig) {
p->numa_hit++;
} else {
p->local_node++;
else
p->other_node++;
- local_irq_restore(flags);
#endif
}
if (free_pages_check(page))
return;
- inc_page_state(pgfree);
kernel_map_pages(page, 1, 0);
pcp = &zone_pcp(zone, get_cpu())->pcp[cold];
local_irq_save(flags);
+ __inc_page_state(pgfree);
list_add(&page->lru, &pcp->list);
pcp->count++;
- if (pcp->count >= pcp->high)
- pcp->count -= free_pages_bulk(zone, pcp->batch, &pcp->list, 0);
+ if (pcp->count >= pcp->high) {
+ free_pages_bulk(zone, pcp->batch, &pcp->list, 0);
+ pcp->count -= pcp->batch;
+ }
local_irq_restore(flags);
put_cpu();
}
* we cheat by calling it from here, in the order > 0 path. Saves a branch
* or two.
*/
-static struct page *
-buffered_rmqueue(struct zone *zone, int order, gfp_t gfp_flags)
+static struct page *buffered_rmqueue(struct zonelist *zonelist,
+ struct zone *zone, int order, gfp_t gfp_flags)
{
unsigned long flags;
struct page *page;
int cold = !!(gfp_flags & __GFP_COLD);
+ int cpu;
again:
- if (order == 0) {
+ cpu = get_cpu();
+ if (likely(order == 0)) {
struct per_cpu_pages *pcp;
- page = NULL;
- pcp = &zone_pcp(zone, get_cpu())->pcp[cold];
+ pcp = &zone_pcp(zone, cpu)->pcp[cold];
local_irq_save(flags);
- if (!pcp->count)
+ if (!pcp->count) {
pcp->count += rmqueue_bulk(zone, 0,
pcp->batch, &pcp->list);
- if (likely(pcp->count)) {
- page = list_entry(pcp->list.next, struct page, lru);
- list_del(&page->lru);
- pcp->count--;
+ if (unlikely(!pcp->count))
+ goto failed;
}
- local_irq_restore(flags);
- put_cpu();
+ page = list_entry(pcp->list.next, struct page, lru);
+ list_del(&page->lru);
+ pcp->count--;
} else {
spin_lock_irqsave(&zone->lock, flags);
page = __rmqueue(zone, order);
- spin_unlock_irqrestore(&zone->lock, flags);
+ spin_unlock(&zone->lock);
+ if (!page)
+ goto failed;
}
- if (page != NULL) {
- BUG_ON(bad_range(zone, page));
- mod_page_state_zone(zone, pgalloc, 1 << order);
- if (prep_new_page(page, order))
- goto again;
+ __mod_page_state_zone(zone, pgalloc, 1 << order);
+ zone_statistics(zonelist, zone, cpu);
+ local_irq_restore(flags);
+ put_cpu();
- if (gfp_flags & __GFP_ZERO)
- prep_zero_page(page, order, gfp_flags);
+ BUG_ON(bad_range(zone, page));
+ if (prep_new_page(page, order))
+ goto again;
- if (order && (gfp_flags & __GFP_COMP))
- prep_compound_page(page, order);
- }
+ if (gfp_flags & __GFP_ZERO)
+ prep_zero_page(page, order, gfp_flags);
+
+ if (order && (gfp_flags & __GFP_COMP))
+ prep_compound_page(page, order);
return page;
+
+failed:
+ local_irq_restore(flags);
+ put_cpu();
+ return NULL;
}
#define ALLOC_NO_WATERMARKS 0x01 /* don't check watermarks at all */
mark = (*z)->pages_high;
if (!zone_watermark_ok(*z, order, mark,
classzone_idx, alloc_flags))
- continue;
+ if (!zone_reclaim_mode ||
+ !zone_reclaim(*z, gfp_mask, order))
+ continue;
}
- page = buffered_rmqueue(*z, order, gfp_mask);
+ page = buffered_rmqueue(zonelist, *z, order, gfp_mask);
if (page) {
- zone_statistics(zonelist, *z);
break;
}
} while (*(++z) != NULL);
*
* The caller may dip into page reserves a bit more if the caller
* cannot run direct reclaim, or if the caller has realtime scheduling
- * policy.
+ * policy or is asking for __GFP_HIGH memory. GFP_ATOMIC requests will
+ * set both ALLOC_HARDER (!wait) and ALLOC_HIGH (__GFP_HIGH).
*/
alloc_flags = ALLOC_WMARK_MIN;
if ((unlikely(rt_task(p)) && !in_interrupt()) || !wait)
cond_resched();
/* We now go into synchronous reclaim */
+ cpuset_memory_pressure_bump();
p->flags |= PF_MEMALLOC;
reclaim_state.reclaimed_slab = 0;
p->reclaim_state = &reclaim_state;
if (page)
goto got_pg;
- out_of_memory(gfp_mask, order);
+ out_of_memory(zonelist, gfp_mask, order);
goto restart;
}
{
int cpu = 0;
- memset(ret, 0, sizeof(*ret));
+ memset(ret, 0, nr * sizeof(unsigned long));
+ cpus_and(*cpumask, *cpumask, cpu_online_map);
cpu = first_cpu(*cpumask);
while (cpu < NR_CPUS) {
unsigned long *in, *out, off;
+ if (!cpu_isset(cpu, *cpumask))
+ continue;
+
in = (unsigned long *)&per_cpu(page_states, cpu);
cpu = next_cpu(cpu, *cpumask);
- if (cpu < NR_CPUS)
+ if (likely(cpu < NR_CPUS))
prefetch(&per_cpu(page_states, cpu));
out = (unsigned long *)ret;
__get_page_state(ret, sizeof(*ret) / sizeof(unsigned long), &mask);
}
-unsigned long __read_page_state(unsigned long offset)
+unsigned long read_page_state_offset(unsigned long offset)
{
unsigned long ret = 0;
int cpu;
- for_each_cpu(cpu) {
+ for_each_online_cpu(cpu) {
unsigned long in;
in = (unsigned long)&per_cpu(page_states, cpu) + offset;
return ret;
}
-void __mod_page_state(unsigned long offset, unsigned long delta)
+void __mod_page_state_offset(unsigned long offset, unsigned long delta)
+{
+ void *ptr;
+
+ ptr = &__get_cpu_var(page_states);
+ *(unsigned long *)(ptr + offset) += delta;
+}
+EXPORT_SYMBOL(__mod_page_state_offset);
+
+void mod_page_state_offset(unsigned long offset, unsigned long delta)
{
unsigned long flags;
- void* ptr;
+ void *ptr;
local_irq_save(flags);
ptr = &__get_cpu_var(page_states);
- *(unsigned long*)(ptr + offset) += delta;
+ *(unsigned long *)(ptr + offset) += delta;
local_irq_restore(flags);
}
-
-EXPORT_SYMBOL(__mod_page_state);
+EXPORT_SYMBOL(mod_page_state_offset);
void __get_zone_counts(unsigned long *active, unsigned long *inactive,
unsigned long *free, struct pglist_data *pgdat)
* Add all populated zones of a node to the zonelist.
*/
static int __init build_zonelists_node(pg_data_t *pgdat,
- struct zonelist *zonelist, int j, int k)
+ struct zonelist *zonelist, int nr_zones, int zone_type)
{
struct zone *zone;
- BUG_ON(k > ZONE_HIGHMEM);
- for (zone = pgdat->node_zones + k; zone >= pgdat->node_zones; zone--) {
+ BUG_ON(zone_type > ZONE_HIGHMEM);
+
+ do {
+ zone = pgdat->node_zones + zone_type;
if (populated_zone(zone)) {
#ifndef CONFIG_HIGHMEM
- BUG_ON(zone - pgdat->node_zones > ZONE_NORMAL);
+ BUG_ON(zone_type > ZONE_NORMAL);
#endif
- zonelist->zones[j++] = zone;
+ zonelist->zones[nr_zones++] = zone;
+ check_highest_zone(zone_type);
}
- }
- return j;
+ zone_type--;
+
+ } while (zone_type >= 0);
+ return nr_zones;
}
static inline int highest_zone(int zone_bits)
*/
static int __init find_next_best_node(int node, nodemask_t *used_node_mask)
{
- int i, n, val;
+ int n, val;
int min_val = INT_MAX;
int best_node = -1;
- for_each_online_node(i) {
- cpumask_t tmp;
+ /* Use the local node if we haven't already */
+ if (!node_isset(node, *used_node_mask)) {
+ node_set(node, *used_node_mask);
+ return node;
+ }
- /* Start from local node */
- n = (node+i) % num_online_nodes();
+ for_each_online_node(n) {
+ cpumask_t tmp;
/* Don't want a node to appear more than once */
if (node_isset(n, *used_node_mask))
continue;
- /* Use the local node if we haven't already */
- if (!node_isset(node, *used_node_mask)) {
- best_node = node;
- break;
- }
-
/* Use the distance array to find the distance */
val = node_distance(node, n);
+ /* Penalize nodes under us ("prefer the next node") */
+ val += (n < node);
+
/* Give preference to headless and unused nodes */
tmp = node_to_cpumask(n);
if (!cpus_empty(tmp))
prev_node = local_node;
nodes_clear(used_mask);
while ((node = find_next_best_node(local_node, &used_mask)) >= 0) {
+ 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;
+
/*
* 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.
*/
- if (node_distance(local_node, node) !=
- node_distance(local_node, prev_node))
+
+ if (distance != node_distance(local_node, prev_node))
node_load[node] += load;
prev_node = node;
load--;
* up by free_all_bootmem() once the early boot process is
* done. Non-atomic initialization, single-pass.
*/
-void __devinit memmap_init_zone(unsigned long size, int nid, unsigned long zone,
+void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone,
unsigned long start_pfn)
{
struct page *page;
unsigned long end_pfn = start_pfn + size;
unsigned long pfn;
- for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) {
+ for (pfn = start_pfn; pfn < end_pfn; pfn++) {
if (!early_pfn_valid(pfn))
continue;
page = pfn_to_page(pfn);
memmap_init_zone((size), (nid), (zone), (start_pfn))
#endif
-static int __devinit zone_batchsize(struct zone *zone)
+static int __cpuinit zone_batchsize(struct zone *zone)
{
int batch;
INIT_LIST_HEAD(&pcp->list);
}
+/*
+ * 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;
+}
+
+
#ifdef CONFIG_NUMA
/*
* Boot pageset table. One per cpu which is going to be used for all
* not check if the processor is online before following the pageset pointer.
* Other parts of the kernel may not check if the zone is available.
*/
-static struct per_cpu_pageset
- boot_pageset[NR_CPUS];
+static struct per_cpu_pageset boot_pageset[NR_CPUS];
/*
* Dynamically allocate memory for the
* per cpu pageset array in struct zone.
*/
-static int __devinit process_zones(int cpu)
+static int __cpuinit process_zones(int cpu)
{
struct zone *zone, *dzone;
for_each_zone(zone) {
- zone->pageset[cpu] = kmalloc_node(sizeof(struct per_cpu_pageset),
+ zone_pcp(zone, cpu) = kmalloc_node(sizeof(struct per_cpu_pageset),
GFP_KERNEL, cpu_to_node(cpu));
- if (!zone->pageset[cpu])
+ if (!zone_pcp(zone, cpu))
goto bad;
- setup_pageset(zone->pageset[cpu], zone_batchsize(zone));
+ setup_pageset(zone_pcp(zone, cpu), zone_batchsize(zone));
+
+ if (percpu_pagelist_fraction)
+ setup_pagelist_highmark(zone_pcp(zone, cpu),
+ (zone->present_pages / percpu_pagelist_fraction));
}
return 0;
for_each_zone(dzone) {
if (dzone == zone)
break;
- kfree(dzone->pageset[cpu]);
- dzone->pageset[cpu] = NULL;
+ kfree(zone_pcp(dzone, cpu));
+ zone_pcp(dzone, cpu) = NULL;
}
return -ENOMEM;
}
static inline void free_zone_pagesets(int cpu)
{
-#ifdef CONFIG_NUMA
struct zone *zone;
for_each_zone(zone) {
zone_pcp(zone, cpu) = NULL;
kfree(pset);
}
-#endif
}
-static int __devinit pageset_cpuup_callback(struct notifier_block *nfb,
+static int __cpuinit pageset_cpuup_callback(struct notifier_block *nfb,
unsigned long action,
void *hcpu)
{
#endif
-static __devinit
+static __meminit
void zone_wait_table_init(struct zone *zone, unsigned long zone_size_pages)
{
int i;
init_waitqueue_head(zone->wait_table + i);
}
-static __devinit void zone_pcp_init(struct zone *zone)
+static __meminit void zone_pcp_init(struct zone *zone)
{
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 */
- zone->pageset[cpu] = &boot_pageset[cpu];
+ zone_pcp(zone, cpu) = &boot_pageset[cpu];
setup_pageset(&boot_pageset[cpu],0);
#else
setup_pageset(zone_pcp(zone,cpu), batch);
zone->name, zone->present_pages, batch);
}
-static __devinit void init_currently_empty_zone(struct zone *zone,
+static __meminit void init_currently_empty_zone(struct zone *zone,
unsigned long zone_start_pfn, unsigned long size)
{
struct pglist_data *pgdat = zone->zone_pgdat;
seq_printf(m,
")"
"\n pagesets");
- for (i = 0; i < ARRAY_SIZE(zone->pageset); i++) {
+ for_each_online_cpu(i) {
struct per_cpu_pageset *pageset;
int j;
return 0;
}
+/*
+ * 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;
+}
+
__initdata int hashdist = HASHDIST_DEFAULT;
#ifdef CONFIG_NUMA
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff