From 94ddf93346cd00bd037df0519820bb45b60b3fb5 Mon Sep 17 00:00:00 2001 From: gatieme Date: Fri, 5 Aug 2016 17:33:09 +0800 Subject: [PATCH] =?UTF-8?q?=E7=80=B9=E5=B1=BD=E6=9D=BD=E6=B5=9C=E5=93=9BEA?= =?UTF-8?q?DME.md...?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../01-description/04-page/README.md | 96 ++++++++++++++++++- 1 file changed, 94 insertions(+), 2 deletions(-) diff --git a/study/kernel/02-memory/01-description/04-page/README.md b/study/kernel/02-memory/01-description/04-page/README.md index afb6167..9211d03 100644 --- a/study/kernel/02-memory/01-description/04-page/README.md +++ b/study/kernel/02-memory/01-description/04-page/README.md @@ -399,6 +399,69 @@ struct page { 其中最后一个flag用于标识page的状态, 这些状态由枚举常量[`enum pageflags`](http://lxr.free-electrons.com/source/include/linux/page-flags.h?v=4.7#L74)定义, 定义在[include/linux/page-flags.h?v=4.7, line 74](http://lxr.free-electrons.com/source/include/linux/page-flags.h?v=4.7#L74). 常用的有如下状态 +```c +enum pageflags { + PG_locked, /* Page is locked. Don't touch. */ + PG_error, + PG_referenced, + PG_uptodate, + PG_dirty, + PG_lru, + PG_active, + PG_slab, + PG_owner_priv_1, /* Owner use. If pagecache, fs may use*/ + PG_arch_1, + PG_reserved, + PG_private, /* If pagecache, has fs-private data */ + PG_private_2, /* If pagecache, has fs aux data */ + PG_writeback, /* Page is under writeback */ + PG_head, /* A head page */ + PG_swapcache, /* Swap page: swp_entry_t in private */ + PG_mappedtodisk, /* Has blocks allocated on-disk */ + PG_reclaim, /* To be reclaimed asap */ + PG_swapbacked, /* Page is backed by RAM/swap */ + PG_unevictable, /* Page is "unevictable" */ +#ifdef CONFIG_MMU + PG_mlocked, /* Page is vma mlocked */ +#endif +#ifdef CONFIG_ARCH_USES_PG_UNCACHED + PG_uncached, /* Page has been mapped as uncached */ +#endif +#ifdef CONFIG_MEMORY_FAILURE + PG_hwpoison, /* hardware poisoned page. Don't touch */ +#endif +#if defined(CONFIG_IDLE_PAGE_TRACKING) && defined(CONFIG_64BIT) + PG_young, + PG_idle, +#endif + __NR_PAGEFLAGS, + + /* Filesystems */ + PG_checked = PG_owner_priv_1, + + /* Two page bits are conscripted by FS-Cache to maintain local caching + * state. These bits are set on pages belonging to the netfs's inodes + * when those inodes are being locally cached. + */ + PG_fscache = PG_private_2, /* page backed by cache */ + + /* XEN */ + /* Pinned in Xen as a read-only pagetable page. */ + PG_pinned = PG_owner_priv_1, + /* Pinned as part of domain save (see xen_mm_pin_all()). */ + PG_savepinned = PG_dirty, + /* Has a grant mapping of another (foreign) domain's page. */ + PG_foreign = PG_owner_priv_1, + + /* SLOB */ + PG_slob_free = PG_private, + + /* Compound pages. Stored in first tail page's flags */ + PG_double_map = PG_private_2, +}; +``` + + | 页面状态 | 描述 | |:-------:|:----:| | PG_locked | 指定了页是否被锁定, 如果该比特未被置位, 说明有使用者正在操作该page, 则内核的其他部分不允许访问该页， 这可以防止内存管理出现竞态条件 | @@ -407,10 +470,10 @@ struct page { | PG_uptodate | 表示page的数据已经与后备存储器是同步的, 即页的数据已经从块设备读取，且没有出错,数据是最新的 | | PG_dirty | 与后备存储器中的数据相比，该page的内容已经被修改. 出于性能能的考虑，页并不在每次改变后立即回写, 因此内核需要使用该标识来表明页面中的数据已经改变, 应该在稍后刷出 | | PG_lru | 表示该page处于LRU链表上， 这有助于实现页面的回收和切换. 内核使用两个最近最少使用(least recently used-LRU)链表来区别活动和不活动页. 如果页在其中一个链表中, 则该位被设置 | -| PG_active | page处于inactive LRU链表, PG_active和PG_referenced一起控制该page的活跃程度，这在内存回收时将会非常有用 | +| PG_active | page处于inactive LRU链表, PG_active和PG_referenced一起控制该page的活跃程度，这在内存回收时将会非常有用
当位于LRU active_list链表上的页面该位被设置, 并在页面移除时清除该位, 它标记了页面是否处于活动状态 | | PG_slab | 该page属于slab分配器 | | PG_onwer_priv_1 | | -| PG_arch_1 | | +| PG_arch_1 | 直接从代码中引用, PG_arch_1是一个体系结构相关的页面状态位, 一般的代码保证了在第一次禁图页面高速缓存时, 该位被清除. 这使得体系结构可以延迟到页面被某个进程映射后， 才可以D-Cache刷盘 | | PG_reserved | 设置该标志，防止该page被交换到swap | | PG_private | 如果page中的private成员非空，则需要设置该标志, 用于I/O的页可使用该字段将页细分为多核缓冲区 | | PG_private_2 | | @@ -457,6 +520,35 @@ TestClearPageXXX(page)：清除page的PG_XXX位，并返回原值 ``` +很多情况下, 需要等待页的状态改变, 然后才能恢复工作. 因此内核提供了两个辅助函数 + +```c +http://lxr.free-electrons.com/source/include/linux/pagemap.h?v=4.7#L495 +/* + * Wait for a page to be unlocked. + * + * This must be called with the caller "holding" the page, + * ie with increased "page->count" so that the page won't + * go away during the wait.. + */ +static inline void wait_on_page_locked(struct page *page) + +// http://lxr.free-electrons.com/source/include/linux/pagemap.h?v=4.7#L504 +/* + * Wait for a page to complete writeback + */ +static inline void wait_on_page_writeback(struct page *page) +``` + +假定内核的一部分在等待一个被锁定的页面, 直至页面被解锁. wait_on_page_locked提供了该功能. 在页面被锁定的情况下, 调用该函数, 内核将进入睡眠. 而在页面解锁后, 睡眠进程会被自动唤醒并继续工作 + +wait_on_page_writeback的工作方式类似, 该函数会等待与页面相关的所有待决回写操作结束, 将页面包含的数据同步到块设备为止. + + +#4 页面映射到管理区 +------- + +