一、Linux伙伴系統分配器
伙伴系統分配器大體上分為兩類。__get_free_pages()類函數返回分配的第一個頁面的線性地址;alloc_pages()類函數返回頁面描述符地址。不管以哪種函數進行分配,最終會調用alloc_pages()進行分配頁面。
為清楚了解其分配制度,先給個伙伴系統數據的存儲框圖
也就是每個order對應一個free_area結構,free_area以不同的類型以鏈表的方式存儲這些內存塊。
二、主分配函數
下面我們來看這個函數(在UMA模式下)
www.2cto.com
#define alloc_pages(gfp_mask, order) \
alloc_pages_node(numa_node_id(), gfp_mask, order)
www.2cto.com
static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask,
unsigned int order)
{
/* Unknown node is current node */
if (nid < 0)
nid = numa_node_id();
return __alloc_pages(gfp_mask, order, node_zonelist(nid, gfp_mask));
}
www.2cto.com
static inline struct page *
__alloc_pages(gfp_t gfp_mask, unsigned int order,
struct zonelist *zonelist)
{
return __alloc_pages_nodemask(gfp_mask, order, zonelist, NULL);
}
上層分配函數__alloc_pages_nodemask()
www.2cto.com
/*
* This is the 'heart' of the zoned buddy allocator.
*/
/*上層分配器運用了各種方式進行*/
struct page *
__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
struct zonelist *zonelist, nodemask_t *nodemask)
{
enum zone_type high_zoneidx = gfp_zone(gfp_mask);
struct zone *preferred_zone;
struct page *page;
/* Convert GFP flags to their corresponding migrate type */
int migratetype = allocflags_to_migratetype(gfp_mask);
gfp_mask &= gfp_allowed_mask;
/*調試用*/
lockdep_trace_alloc(gfp_mask);
/*如果__GFP_WAIT標志設置了,需要等待和重新調度*/
might_sleep_if(gfp_mask & __GFP_WAIT);
/*沒有設置對應的宏*/
if (should_fail_alloc_page(gfp_mask, order))
return NULL;
/*
* Check the zones suitable for the gfp_mask contain at least one
* valid zone. It's possible to have an empty zonelist as a result
* of GFP_THISNODE and a memoryless node
*/
if (unlikely(!zonelist->_zonerefs->zone))
return NULL;
/* The preferred zone is used for statistics later */
/* 英文注釋所說*/
first_zones_zonelist(zonelist, high_zoneidx, nodemask, &preferred_zone);
if (!preferred_zone)
return NULL;
/* First allocation attempt */
/*從pcp和伙伴系統中正常的分配內存空間*/
page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, nodemask, order,
zonelist, high_zoneidx, ALLOC_WMARK_LOW|ALLOC_CPUSET,
preferred_zone, migratetype);
if (unlikely(!page))/*如果上面沒有分配到空間,調用下面函數慢速分配,允許等待和回收*/
page = __alloc_pages_slowpath(gfp_mask, order,
zonelist, high_zoneidx, nodemask,
preferred_zone, migratetype);
/*調試用*/
trace_mm_page_alloc(page, order, gfp_mask, migratetype);
return page;
}
三、從pcp和伙伴系統中正常的分配內存空間
函數get_page_from_freelist()
www.2cto.com
/*
* get_page_from_freelist goes through the zonelist trying to allocate
* a page.
*/
/*為分配制定內存空間,遍歷每個zone*/
static struct page *
get_page_from_freelist(gfp_t gfp_mask, nodemask_t *nodemask, unsigned int order,
struct zonelist *zonelist, int high_zoneidx, int alloc_flags,
struct zone *preferred_zone, int migratetype)
{
struct zoneref *z;
struct page *page = NULL;
int classzone_idx;
struct zone *zone;
nodemask_t *allowednodes = NULL;/* zonelist_cache approximation */
int zlc_active = 0; /* set if using zonelist_cache */
int did_zlc_setup = 0; /* just call zlc_setup() one time */
/*zone對應的下標*/
classzone_idx = zone_idx(preferred_zone);
zonelist_scan:
/*
* Scan zonelist, looking for a zone with enough free.
* See also cpuset_zone_allowed() comment in kernel/cpuset.c.
*/
/*遍歷每個zone,進行分配*/
for_each_zone_zonelist_nodemask(zone, z, zonelist,
/*在UMA模式下不成立*/ high_zoneidx, nodemask) {
if (NUMA_BUILD && zlc_active &&
!zlc_zone_worth_trying(zonelist, z, allowednodes))
continue;
if ((alloc_flags & ALLOC_CPUSET) &&
!cpuset_zone_allowed_softwall(zone, gfp_mask))
goto try_next_zone;
BUILD_BUG_ON(ALLOC_NO_WATERMARKS < NR_WMARK);
/*需要關注水位*/
if (!(alloc_flags & ALLOC_NO_WATERMARKS)) {
unsigned long mark;
int ret;
/*從flags中取的mark*/
mark = zone->watermark[alloc_flags & ALLOC_WMARK_MASK];
/*如果水位正常,從本zone中分配*/
if (zone_watermark_ok(zone, order, mark,
classzone_idx, alloc_flags))
goto try_this_zone;
if (zone_reclaim_mode == 0)/*如果上面檢查的水位低於正常值,且沒有設置頁面回收值*/
goto this_zone_full;
/*在UMA模式下下面函數直接返回0*/
ret = zone_reclaim(zone, gfp_mask, order);
switch (ret) {
case ZONE_RECLAIM_NOSCAN:
/* did not scan */
goto try_next_zone;
case ZONE_RECLAIM_FULL:
/* scanned but unreclaimable */
goto this_zone_full;
default:
/* did we reclaim enough */
if (!zone_watermark_ok(zone, order, mark,
classzone_idx, alloc_flags))
goto this_zone_full;
}
}
try_this_zone:/*本zone正常水位*/
/*先從pcp中分配,然後不行的話再從伙伴系統中分配*/
page = buffered_rmqueue(preferred_zone, zone, order,
gfp_mask, migratetype);
if (page)
break;
this_zone_full:
if (NUMA_BUILD)/*UMA模式為0*/
zlc_mark_zone_full(zonelist, z);
try_next_zone:
if (NUMA_BUILD && !did_zlc_setup && nr_online_nodes > 1) {
/*
* we do zlc_setup after the first zone is tried but only
* if there are multiple nodes make it worthwhile
*/
allowednodes = zlc_setup(zonelist, alloc_flags);
zlc_active = 1;
did_zlc_setup = 1;
}
}
if (unlikely(NUMA_BUILD && page == NULL && zlc_active)) {
/* Disable zlc cache for second zonelist scan */
zlc_active = 0;
goto zonelist_scan;
}
return page;/*返回頁面*/
}
主分配函數
www.2cto.com
/*
* 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.
*/
/*先考慮從pcp中分配空間,當order大於0時再考慮從伙伴系統中分配*/
static inline
struct page *buffered_rmqueue(struct zone *preferred_zone,
struct zone *zone, int order, gfp_t gfp_flags,
int migratetype)
{
unsigned long flags;
struct page *page;
int cold = !!(gfp_flags & __GFP_COLD);/*如果分配參數指定了__GFP_COLD標志,則設置cold標志*/
int cpu;
again:
cpu = get_cpu();
if (likely(order == 0)) {/*分配一個頁面時,使用pcp*/
struct per_cpu_pages *pcp;
struct list_head *list;
/*找到zone對應的pcp*/
pcp = &zone_pcp(zone, cpu)->pcp;
list = &pcp->lists[migratetype];/*pcp中對應類型的list*/
/* 這裡需要關中斷,因為內存回收過程可能發送核間中斷,強制每個核從每CPU
緩存中釋放頁面。而且中斷處理函數也會分配單頁。*/
local_irq_save(flags);
if (list_empty(list)) {/*如果pcp中沒有頁面,需要補充*/
/*從伙伴系統中獲得batch個頁面
batch為一次分配的頁面數*/
pcp->count += rmqueue_bulk(zone, 0,
pcp->batch, list,
migratetype, cold);
/*如果鏈表仍然為空,申請失敗返回*/
if (unlikely(list_empty(list)))
goto failed;
}
/* 如果分配的頁面不需要考慮硬件緩存(注意不是每CPU頁面緩存)
,則取出鏈表的最後一個節點返回給上層*/
if (cold)
page = list_entry(list->prev, struct page, lru);
else/* 如果要考慮硬件緩存,則取出鏈表的第一個頁面,這個頁面是最近剛釋放到每CPU
緩存的,緩存熱度更高*/
page = list_entry(list->next, struct page, lru);
list_del(&page->lru);/*從pcp中脫離*/
pcp->count--;/*pcp計數減一*/
}
else {/*當order為大於1時,不從pcp中分配,直接考慮從伙伴系統中分配*/
if (unlikely(gfp_flags & __GFP_NOFAIL)) {
/*
* __GFP_NOFAIL is not to be used in new code.
*
* All __GFP_NOFAIL callers should be fixed so that they
* properly detect and handle allocation failures.
*
* We most definitely don't want callers attempting to
* allocate greater than order-1 page units with
* __GFP_NOFAIL.
*/
WARN_ON_ONCE(order > 1);
}
/* 關中斷,並獲得管理區的鎖*/
spin_lock_irqsave(&zone->lock, flags);
/*從伙伴系統中相應類型的相應鏈表中分配空間*/
page = __rmqueue(zone, order, migratetype);
/* 已經分配了1 << order個頁面,這裡進行管理區空閒頁面統計計數*/
__mod_zone_page_state(zone, NR_FREE_PAGES, -(1 << order));
spin_unlock(&zone->lock);/* 這裡僅僅打開自旋鎖,待後面統計計數設置完畢後再開中斷*/
if (!page)
goto failed;
}
/*事件統計計數,調試*/
__count_zone_vm_events(PGALLOC, zone, 1 << order);
zone_statistics(preferred_zone, zone);
local_irq_restore(flags);/*恢復中斷*/
put_cpu();
VM_BUG_ON(bad_range(zone, page));
/* 這裡進行安全性檢查,並進行一些善後工作。
如果頁面標志破壞,返回的頁面出現了問題,則返回試圖分配其他頁面*/
if (prep_new_page(page, order, gfp_flags))
goto again;
return page;
failed:
local_irq_restore(flags);
put_cpu();
return NULL;
}
3.1 pcp緩存補充
從伙伴系統中獲得batch個頁面,batch為一次分配的頁面數rmqueue_bulk()函數。
www.2cto.com
/*
* 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.
*/
/*該函數返回的是1<<order個頁面,但是在pcp
處理中調用,其他地方沒看到,order為0
也就是說返回的是頁面數,加入的鏈表為
對應調用pcp的鏈表*/
static int rmqueue_bulk(struct zone *zone, unsigned int order,
unsigned long count, struct list_head *list,
int migratetype, int cold)
{
int i;
spin_lock(&zone->lock);/* 上層函數已經關了中斷,這裡需要操作管理區,獲取管理區的自旋鎖*/
for (i = 0; i < count; ++i) {/* 重復指定的次數,從伙伴系統中分配頁面*/
/* 從伙伴系統中取出頁面*/
struct page *page = __rmqueue(zone, order, migratetype);
if (unlikely(page == NULL))/*分配失敗*/
break;
/*
* Split buddy pages returned by expand() are received here
* in physical page order. The page is added to the callers and
* list and the list head then moves forward. From the callers
* perspective, the linked list is ordered by page number in
* some conditions. This is useful for IO devices that can
* merge IO requests if the physical pages are ordered
* properly.
*/
if (likely(cold == 0))/*根據調用者的要求,將頁面放到每CPU緩存鏈表的頭部或者尾部*/
list_add(&page->lru, list);
else
list_add_tail(&page->lru, list);
set_page_private(page, migratetype);/*設置private屬性為頁面的遷移類型*/
list = &page->lru;
}
/*遞減管理區的空閒頁面計數*/
__mod_zone_page_state(zone, NR_FREE_PAGES, -(i << order));
spin_unlock(&zone->lock);/*釋放管理區的子璇鎖*/
return i;
}
3.2 從伙伴系統中取出頁面
__rmqueue()函數
www.2cto.com
/*
* Do the hard work of removing an element from the buddy allocator.
* Call me with the zone->lock already held.
*/
/*采用兩種范式試著分配order個page*/
static struct page *__rmqueue(struct zone *zone, unsigned int order,
int migratetype)
{
struct page *page;
retry_reserve:
/*從指定order開始從小到達遍歷,優先從指定的遷移類型鏈表中分配頁面*/
page = __rmqueue_smallest(zone, order, migratetype);
/*
* 如果滿足以下兩個條件,就從備用鏈表中分配頁面:
* 快速流程沒有分配到頁面,需要從備用遷移鏈表中分配.
* 當前不是從保留的鏈表中分配.因為保留的鏈表是最後可用的鏈表,
* 不能從該鏈表分配的話,說明本管理區真的沒有可用內存了.
*/
if (unlikely(!page) && migratetype != MIGRATE_RESERVE) {
/*order從大到小遍歷,從備用鏈表中分配頁面*/
page = __rmqueue_fallback(zone, order, migratetype);
/*
* Use MIGRATE_RESERVE rather than fail an allocation. goto
* is used because __rmqueue_smallest is an inline function
* and we want just one call site
*/
if (!page) {/* 備用鏈表中沒有分配到頁面,從保留鏈表中分配頁面了*/
migratetype = MIGRATE_RESERVE;
goto retry_reserve;/* 跳轉到retry_reserve,從保留的鏈表中分配頁面*/
}
}
/*調試代碼*/
trace_mm_page_alloc_zone_locked(page, order, migratetype);
return page;
}
3.2.1 從指定的遷移類型鏈表中分配頁面
從指定order開始從小到達遍歷,優先從指定的遷移類型鏈表中分配頁面__rmqueue_smallest(zone, order, migratetype);
www.2cto.com
/*
* Go through the free lists for the given migratetype and remove
* the smallest available page from the freelists
*/
/*從給定的order開始,從小到大遍歷;
找到後返回頁面基址,合並分割後的空間*/
static inline
struct page *__rmqueue_smallest(struct zone *zone, unsigned int order,
int migratetype)
{
unsigned int current_order;
struct free_area * area;
struct page *page;
/* Find a page of the appropriate size in the preferred list */
for (current_order = order; current_order < MAX_ORDER; ++current_order) {
area = &(zone->free_area[current_order]);/*得到指定order的area*/
/*如果area指定類型的伙伴系統鏈表為空*/
if (list_empty(&area->free_list[migratetype]))
continue;/*查找下一個order*/
/*對應的鏈表不空,得到鏈表中數據*/
page = list_entry(area->free_list[migratetype].next,
struct page, lru);
list_del(&page->lru);/*從伙伴系統中刪除;*/
rmv_page_order(page);/*移除page中order的變量*/
area->nr_free--;/*空閒塊數減一*/
/*拆分、合並*/
expand(zone, page, order, current_order, area, migratetype);
return page;
}
return NULL;
}
伙伴系統內存塊拆分和合並
看一個輔助函數,用於伙伴系統中內存塊的拆分、合並
www.2cto.com
/*
* 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
*/
/*此函數主要用於下面這種情況:
分配函數從high中分割出去了low大小的內存;
然後要將high留下的內存塊合並放到伙伴系統中;*/
static inline void expand(struct zone *zone, struct page *page,
int low, int high, struct free_area *area,
int migratetype)
{
unsigned long size = 1 << high;
while (high > low) {/*因為去掉了low的大小,所以最後肯定剩下的
是low的大小(2的指數運算)*/
area--;/*減一到order減一的area*/
high--;/*order減一*/
size >>= 1;/*大小除以2*/
VM_BUG_ON(bad_range(zone, &page[size]));
/*加到指定的伙伴系統中*/
list_add(&page[size].lru, &area->free_list[migratetype]);
area->nr_free++;/*空閒塊加一*/
set_page_order(&page[size], high);/*設置相關order*/
}
}
3.2.2 從備用鏈表中分配頁面
www.2cto.com
/* Remove an element from the buddy allocator from the fallback list */
static inline struct page *
__rmqueue_fallback(struct zone *zone, int order, int start_migratetype)
{
struct free_area * area;
int current_order;
struct page *page;
int migratetype, i;
/* Find the largest possible block of pages in the other list */
/* 從最高階搜索,這樣可以盡量的將其他遷移列表中的大塊分割,避免形成過多的碎片*/
for (current_order = MAX_ORDER-1; current_order >= order;
--current_order) {
for (i = 0; i < MIGRATE_TYPES - 1; i++) {
/*回調到下一個migratetype*/
migratetype = fallbacks[start_migratetype][i];
/* MIGRATE_RESERVE handled later if necessary */
/* 本函數不處理MIGRATE_RESERVE類型的遷移鏈表,如果本函數返回NULL,
則上層函數直接從MIGRATE_RESERVE中分配*/
if (migratetype == MIGRATE_RESERVE)
continue;/*訪問下一個類型*/
area = &(zone->free_area[current_order]);
/*如果指定order和類型的鏈表為空*/
if (list_empty(&area->free_list[migratetype]))
continue;/*訪問下一個類型*/
/*得到指定類型和order的頁面基址*/
page = list_entry(area->free_list[migratetype].next,
struct page, lru);
area->nr_free--;/*空閒塊數減一*/
/*
* If breaking a large block of pages, move all free
* pages to the preferred allocation list. If falling
* back for a reclaimable kernel allocation, be more
* agressive about taking ownership of free pages
*/
if (unlikely(current_order >= (pageblock_order >> 1)) ||/* 要分割的頁面是一個大頁面,則將整個頁面全部遷移到當前遷移類型的鏈表中,
這樣可以避免過多的碎片*/
start_migratetype == MIGRATE_RECLAIMABLE ||/* 目前分配的是可回收頁面,這類頁面有突發的特點,將頁面全部遷移到可回收鏈表中,
可以避免將其他遷移鏈表分割成太多的碎片*/
page_group_by_mobility_disabled) {/* 指定了遷移策略,總是將被分割的頁面遷移*/
unsigned long pages;
/*移動到先前類型的伙伴系統中*/
pages = move_freepages_block(zone, page,
start_migratetype);
/* Claim the whole block if over half of it is free */
/* pages是移動的頁面數,如果可移動的頁面數量較多,
則將整個大內存塊的遷移類型修改*/
if (pages >= (1 << (pageblock_order-1)) ||
page_group_by_mobility_disabled)
/*設置頁面標示*/
set_pageblock_migratetype(page,
start_migratetype);
migratetype = start_migratetype;
}
/* Remove the page from the freelists */
list_del(&page->lru);
rmv_page_order(page);
/* Take ownership for orders >= pageblock_order */
if (current_order >= pageblock_order)//大於pageblock_order的部分設置相應標示
/*這個不太可能,因為pageblock_order為10*/
change_pageblock_range(page, current_order,
start_migratetype);
/*拆分和合並*/
expand(zone, page, order, current_order, area, migratetype);
trace_mm_page_alloc_extfrag(page, order, current_order,
start_migratetype, migratetype);
return page;
}
}
return NULL;
}
備用鏈表
www.2cto.com
/*
* This array describes the order lists are fallen back to when
* the free lists for the desirable migrate type are depleted
*/
/*指定類型的鏈表為空時,這個數組規定
回調的到那個類型的鏈表*/
static int fallbacks[MIGRATE_TYPES][MIGRATE_TYPES-1] = {
[MIGRATE_UNMOVABLE] = { MIGRATE_RECLAIMABLE, MIGRATE_MOVABLE, MIGRATE_RESERVE },
[MIGRATE_RECLAIMABLE] = { MIGRATE_UNMOVABLE, MIGRATE_MOVABLE, MIGRATE_RESERVE },
[MIGRATE_MOVABLE] = { MIGRATE_RECLAIMABLE, MIGRATE_UNMOVABLE, MIGRATE_RESERVE },
[MIGRATE_RESERVE] = { MIGRATE_RESERVE, MIGRATE_RESERVE, MIGRATE_RESERVE }, /* Never used */
};
移動到指定類型的伙伴系統中
www.2cto.com
/*將指定區域段的頁面移動到指定類型的
伙伴系統中,其實就是將頁面的類型做了
更改,但是是采用移動的方式
功能和上面函數類似,但是要求以
頁面塊方式對其*/
static int move_freepages_block(struct zone *zone, struct page *page,
int migratetype)
{
unsigned long start_pfn, end_pfn;
struct page *start_page, *end_page;
/*如下是對齊操作,其中變量pageblock_nr_pages為MAX_ORDER-1*/
start_pfn = page_to_pfn(page);
start_pfn = start_pfn & ~(pageblock_nr_pages-1);
start_page = pfn_to_page(start_pfn);
end_page = start_page + pageblock_nr_pages - 1;
end_pfn = start_pfn + pageblock_nr_pages - 1;
/* Do not cross zone boundaries */
if (start_pfn < zone->zone_start_pfn)
start_page = page;
/*結束邊界檢查*/
if (end_pfn >= zone->zone_start_pfn + zone->spanned_pages)
return 0;
/*調用上面函數*/
return move_freepages(zone, start_page, end_page, migratetype);
}
www.2cto.com
/*
* Move the free pages in a range to the free lists of the requested type.
* Note that start_page and end_pages are not aligned on a pageblock
* boundary. If alignment is required, use move_freepages_block()
*/
/*將指定區域段的頁面移動到指定類型的
伙伴系統中,其實就是將頁面的類型做了 更改,但是是采用移動的方式*/
static int move_freepages(struct zone *zone,
struct page *start_page, struct page *end_page,
int migratetype)
{
struct page *page;
unsigned long order;
int pages_moved = 0;
#ifndef CONFIG_HOLES_IN_ZONE
/*
* page_zone is not safe to call in this context when
* CONFIG_HOLES_IN_ZONE is set. This bug check is probably redundant
* anyway as we check zone boundaries in move_freepages_block().
* Remove at a later date when no bug reports exist related to
* grouping pages by mobility
*/
BUG_ON(page_zone(start_page) != page_zone(end_page));
#endif
for (page = start_page; page <= end_page;) {
/* Make sure we are not inadvertently changing nodes */
VM_BUG_ON(page_to_nid(page) != zone_to_nid(zone));
if (!pfn_valid_within(page_to_pfn(page))) {
page++;
continue;
}
if (!PageBuddy(page)) {
page++;
continue;
}
order = page_order(page);
list_del(&page->lru);/*將頁面塊從原來的伙伴系統鏈表*/
/*中刪除,注意,這裡不是一個頁面
*而是以該頁面的伙伴塊*/
list_add(&page->lru,/*添加到指定order和類型下的伙伴系統鏈表*/
&zone->free_area[order].free_list[migratetype]);
page += 1 << order;/*移動頁面數往上定位*/
pages_moved += 1 << order;/*移動的頁面數*/
}
return pages_moved;
}
四、慢速分配,允許等待和回收
www.2cto.com
/**
* 當無法快速分配頁面時,如果調用者允許等待
,則通過本函數進行慢速分配。
* 此時允許進行內存回收。
*/
static inline struct page *
__alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
struct zonelist *zonelist, enum zone_type high_zoneidx,
nodemask_t *nodemask, struct zone *preferred_zone,
int migratetype)
{
const gfp_t wait = gfp_mask & __GFP_WAIT;
struct page *page = NULL;
int alloc_flags;
unsigned long pages_reclaimed = 0;
unsigned long did_some_progress;
struct task_struct *p = current;
/*
* In the slowpath, we sanity check order to avoid ever trying to
* reclaim >= MAX_ORDER areas which will never succeed. Callers may
* be using allocators in order of preference for an area that is
* too large.
*//*參數合法性檢查*/
if (order >= MAX_ORDER) {
WARN_ON_ONCE(!(gfp_mask & __GFP_NOWARN));
return NULL;
}
/*
* GFP_THISNODE (meaning __GFP_THISNODE, __GFP_NORETRY and
* __GFP_NOWARN set) should not cause reclaim since the subsystem
* (f.e. slab) using GFP_THISNODE may choose to trigger reclaim
* using a larger set of nodes after it has established that the
* allowed per node queues are empty and that nodes are
* over allocated.
*/
/**
* 調用者指定了GFP_THISNODE標志,表示不能進行內存回收。
* 上層調用者應當在指定了GFP_THISNODE失敗後,使用其他標志進行分配。
*/
if (NUMA_BUILD && (gfp_mask & GFP_THISNODE) == GFP_THISNODE)
goto nopage;
restart:/*如果調用者沒有禁止kswapd,則喚醒該線程進行內存回收。*/
wake_all_kswapd(order, zonelist, high_zoneidx);
/*
* 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.
*/
/*根據分配標志確定內部標志,主要是用於水線*/
alloc_flags = gfp_to_alloc_flags(gfp_mask);
/**
* 與快速分配流程相比,這裡的分配標志使用了低的水線。
* 在進行內存回收操作前,我們使用低水線再嘗試分配一下。
* 當然,不管是否允許ALLOC_NO_WATERMARKS標志,我們都將它清除。
*/
/* This is the last chance, in general, before the goto nopage. */
page = get_page_from_freelist(gfp_mask, nodemask, order, zonelist,
high_zoneidx, alloc_flags & ~ALLOC_NO_WATERMARKS,
preferred_zone, migratetype);
if (page)/*分配成功,找到頁面*/
goto got_pg;
rebalance:
/* Allocate without watermarks if the context allows */
/* 某些上下文,如內存回收進程及被殺死的任務,都允許它完全突破水線的限制分配內存。*/
if (alloc_flags & ALLOC_NO_WATERMARKS) {
page = __alloc_pages_high_priority(gfp_mask, order,
zonelist, high_zoneidx, nodemask,
preferred_zone, migratetype);
if (page))/* 在不考慮水線的情況下,分配到了內存*/
goto got_pg;
}
/* Atomic allocations - we can't balance anything */
/* 調用者希望原子分配內存,此時不能等待內存回收,返回NULL */
if (!wait)
goto nopage;
/* Avoid recursion of direct reclaim */
/* 調用者本身就是內存回收進程,不能進入後面的內存回收處理流程,否則死鎖*/
if (p->flags & PF_MEMALLOC)
goto nopage;
/* Avoid allocations with no watermarks from looping endlessly */
/**
* 當前線程正在被殺死,它可以完全突破水線分配內存。這裡向上層返回NULL,是為了避免系統進入死循環。
* 當然,如果上層調用不允許失敗,則死循環繼續分配,等待其他線程釋放一點點內存。
*/
if (test_thread_flag(TIF_MEMDIE) && !(gfp_mask & __GFP_NOFAIL))
goto nopage;
/* Try direct reclaim and then allocating */
/**
* 直接在內存分配上下文中進行內存回收操作。
*/
page = __alloc_pages_direct_reclaim(gfp_mask, order,
zonelist, high_zoneidx,
nodemask,
alloc_flags, preferred_zone,
migratetype, &did_some_progress);
if (page))/* 慶幸,回收了一些內存後,滿足了上層分配需求*/
goto got_pg;
/*
* If we failed to make any progress reclaiming, then we are
* running out of options and have to consider going OOM
*/
/* 內存回收過程沒有回收到內存,系統真的內存不足了*/
if (!did_some_progress) {
/**
* 調用者不是文件系統的代碼,允許進行文件系統操作,並且允許重試。
* 這裡需要__GFP_FS標志可能是進入OOM流程後會殺進程或進入panic,需要文件操作。
*/
if ((gfp_mask & __GFP_FS) && !(gfp_mask & __GFP_NORETRY)) {
if (oom_killer_disabled)/* 系統禁止了OOM,向上層返回NULL */
goto nopage;
/**
* 殺死其他進程後再嘗試分配內存
*/
page = __alloc_pages_may_oom(gfp_mask, order,
zonelist, high_zoneidx,
nodemask, preferred_zone,
migratetype);
if (page)
goto got_pg;
/*
* The OOM killer does not trigger for high-order
* ~__GFP_NOFAIL allocations so if no progress is being
* made, there are no other options and retrying is
* unlikely to help.
*/)/* 要求的頁面數量較多,再試意義不大*/
if (order > PAGE_ALLOC_COSTLY_ORDER &&
!(gfp_mask & __GFP_NOFAIL))
goto nopage;
goto restart;
}
}
/* Check if we should retry the allocation */
/* 內存回收過程回收了一些內存,接下來判斷是否有必要繼續重試*/
pages_reclaimed += did_some_progress;
if (should_alloc_retry(gfp_mask, order, pages_reclaimed)) {
/* Wait for some write requests to complete then retry */
congestion_wait(BLK_RW_ASYNC, HZ/50);
goto rebalance;
}
nopage:
/* 內存分配失敗了,打印內存分配失敗的警告*/
if (!(gfp_mask & __GFP_NOWARN) && printk_ratelimit()) {
printk(KERN_WARNING "%s: page allocation failure."
" order:%d, mode:0x%x\n",
p->comm, order, gfp_mask);
dump_stack();
show_mem();
}
return page;
got_pg:
/* 運行到這裡,說明成功分配了內存,這裡進行內存檢測調試*/
if (kmemcheck_enabled)
kmemcheck_pagealloc_alloc(page, order, gfp_mask);
return page;
}
總結:Linux伙伴系統主要分配流程為
正常非配(或叫快速分配)流程:
1,如果分配的是單個頁面,考慮從per CPU緩存中分配空間,如果緩存中沒有頁面,從伙伴系統中提取頁面做補充。
2,分配多個頁面時,從指定類型中分配,如果指定類型中沒有足夠的頁面,從備用類型鏈表中分配。最後會試探保留類型鏈表。
慢速(允許等待和頁面回收)分配:
3,當上面兩種分配方案都不能滿足要求時,考慮頁面回收、殺死進程等操作後在試