linux内存管理（八）- 反向映射RMAPseo优化

linux内存管理（八）- 反向映射RMAP

news/2024/10/5 5:13:54

这里有一篇博客讲的不错。linux内存管理笔记(三十八）----反向映射_linux 反向映射-CSDN博客

页表是把虚拟地址映射到物理页面，但是如何根据一个物理页找到所有映射它的pte呢？答案是用反向映射Reverse Mapping（RMAP)。这在页面回收中很有用。回收页面需要将到物理页的映射断开（改一下pte），前提是找到所有映射的pte，那就必须在page结构中留下线索，这就是mapping字段。

在讲page结构时我们提到过mapping可以指向匿名页的anon_vma结构，这个结构是反向映射的关键之一。

struct anon_vma {struct anon_vma *root;        /* Root of this anon_vma tree */struct rw_semaphore rwsem;    /* W: modification, R: walking the list */
    atomic_t refcount;
    unsigned long num_children;/* Count of VMAs whose ->anon_vma pointer points to this object. */unsigned long num_active_vmas;struct anon_vma *parent;    /* Parent of this anon_vma */struct rb_root_cached rb_root;  //avc结构会链接到这棵rbtree上
};

除了anon_vma（AV）还有一个AVC（anon_vma_chain)结构。

/** The copy-on-write semantics of fork mean that an anon_vma* can become associated with multiple processes. Furthermore,* each child process will have its own anon_vma, where new* pages for that process are instantiated.** This structure allows us to find the anon_vmas associated* with a VMA, or the VMAs associated with an anon_vma.* The "same_vma" list contains the anon_vma_chains linking* all the anon_vmas associated with this VMA.* The "rb" field indexes on an interval tree the anon_vma_chains* which link all the VMAs associated with this anon_vma.*/
struct anon_vma_chain {struct vm_area_struct *vma;struct anon_vma *anon_vma;struct list_head same_vma;   /* locked by mmap_lock & page_table_lock */struct rb_node rb;            /* locked by anon_vma->rwsem */unsigned long rb_subtree_last;
#ifdef CONFIG_DEBUG_VM_RBunsigned long cached_vma_start, cached_vma_last;
#endif
};

看注释可知AVC可以通过rb_node链接到anon_vma的rbtree上，这个rbtree会链接所有与AV相关的VMA。也可以通过same_vma链表链接所有与该VMA相关的AV。AVC是anon_vma和vma的枢纽，可以让三者相互找到另外两方。

上图是一个简单的情形，描述三者之间的关系，这对理解后面反向映射的应用很重要。

反向映射的应用。

使用反向映射最常见的是在回收页面，try_to_unmap是其重要函数。

void try_to_unmap(struct folio *folio, enum ttu_flags flags)
{struct rmap_walk_control rwc = {.rmap_one = try_to_unmap_one,.arg = (void *)flags,.done = folio_not_mapped,.anon_lock = folio_lock_anon_vma_read,};if (flags & TTU_RMAP_LOCKED)rmap_walk_locked(folio, &rwc);elsermap_walk(folio, &rwc);
}

rmap_walk

void rmap_walk(struct folio *folio, struct rmap_walk_control *rwc)
{if (unlikely(folio_test_ksm(folio)))rmap_walk_ksm(folio, rwc);else if (folio_test_anon(folio))rmap_walk_anon(folio, rwc, false);elsermap_walk_file(folio, rwc, false);
}

只看rmap_walk_anon

static void rmap_walk_anon(struct folio *folio,struct rmap_walk_control *rwc, bool locked)
{struct anon_vma *anon_vma;pgoff_t pgoff_start, pgoff_end;struct anon_vma_chain *avc;if (locked) {anon_vma = folio_anon_vma(folio);/* anon_vma disappear under us? */VM_BUG_ON_FOLIO(!anon_vma, folio);} else {
        //从folio中获取anon_vmaanon_vma = rmap_walk_anon_lock(folio, rwc);}if (!anon_vma)return;//page->index是page在file或内存区域中的偏移，单位是pagepgoff_start = folio_pgoff(folio);
    //得到复合页的末尾indexpgoff_end = pgoff_start + folio_nr_pages(folio) - 1;
    //遍历avcanon_vma_interval_tree_foreach(avc, &anon_vma->rb_root,pgoff_start, pgoff_end) {struct vm_area_struct *vma = avc->vma;
        //得到page的虚拟地址unsigned long address = vma_address(&folio->page, vma);VM_BUG_ON_VMA(address == -EFAULT, vma);cond_resched();if (rwc->invalid_vma && rwc->invalid_vma(vma, rwc->arg))continue;//使用回调函数rmap_one断开虚拟地址对应的pteif (!rwc->rmap_one(folio, vma, address, rwc->arg))break;if (rwc->done && rwc->done(folio))break;}if (!locked)anon_vma_unlock_read(anon_vma);
}