好记性不如铅笔头

kernel, linux, 操作系统

Linux高端内存映射学习笔记

最近在看《深入理解LINUX内核》的高端内存映射,这里笔记下代码的实现吧。

highmem.c

/*
 * High memory handling common code and variables.
 *
 * (C) 1999 Andrea Arcangeli, SuSE GmbH, andrea@suse.de
 *          Gerhard Wichert, Siemens AG, Gerhard.Wichert@pdb.siemens.de
 *
 *
 * Redesigned the x86 32-bit VM architecture to deal with
 * 64-bit physical space. With current x86 CPUs this
 * means up to 64 Gigabytes physical RAM.
 *
 * Rewrote high memory support to move the page cache into
 * high memory. Implemented permanent (schedulable) kmaps
 * based on Linus' idea.
 *
 * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
 */

#include <linux/mm.h>
#include <linux/module.h>
#include <linux/swap.h>
#include <linux/bio.h>
#include <linux/pagemap.h>
#include <linux/mempool.h>
#include <linux/blkdev.h>
#include <linux/init.h>
#include <linux/hash.h>
#include <linux/highmem.h>
#include <asm/tlbflush.h>

static mempool_t *page_pool, *isa_page_pool;

static void *page_pool_alloc(int gfp_mask, void *data)
{
	int gfp = gfp_mask | (int) (long) data;

	return alloc_page(gfp);
}

static void page_pool_free(void *page, void *data)
{
	__free_page(page);
}

/*
 * Virtual_count is not a pure "count".
 *  0 means that it is not mapped, and has not been mapped
 *    since a TLB flush - it is usable.
 *  1 means that there are no users, but it has been mapped
 *    since the last TLB flush - so we can't use it.
 *  n means that there are (n-1) current users of it.
 */
#ifdef CONFIG_HIGHMEM
static int pkmap_count[LAST_PKMAP];
static unsigned int last_pkmap_nr;
static  __cacheline_aligned_in_smp DEFINE_SPINLOCK(kmap_lock);

pte_t * pkmap_page_table;

static DECLARE_WAIT_QUEUE_HEAD(pkmap_map_wait);

/*
刷新映射表,将计数为1的全部干掉,通过unmap可以发现,该函数不会主动彻底
释放映射,只是会唤醒等待队列,然后等待任务呗唤醒,通过循环进入本函数
*/
static void flush_all_zero_pkmaps(void)
{
	int i;

	flush_cache_kmaps();

	for (i = 0; i < LAST_PKMAP; i++) {
		struct page *page;

		/*
		 * zero means we don't have anything to do,
		 * >1 means that it is still in use. Only
		 * a count of 1 means that it is free but
		 * needs to be unmapped
		 */
		if (pkmap_count[i] != 1)
			continue;
		pkmap_count[i] = 0;

		/* sanity check */
		if (pte_none(pkmap_page_table[i]))
			BUG();

		/*
		 * Don't need an atomic fetch-and-clear op here;
		 * no-one has the page mapped, and cannot get at
		 * its virtual address (and hence PTE) without first
		 * getting the kmap_lock (which is held here).
		 * So no dangers, even with speculative execution.
		 */
		page = pte_page(pkmap_page_table[i]);
		pte_clear(&pkmap_page_table[i]);

		/* 释放内存映射 */
		set_page_address(page, NULL);
	}
	flush_tlb_kernel_range(PKMAP_ADDR(0), PKMAP_ADDR(LAST_PKMAP));
}

/*

*/
static inline unsigned long map_new_virtual(struct page *page)
{
	unsigned long vaddr;
	int count;

start:
	count = LAST_PKMAP;
	/* Find an empty entry */
	/*
	首先寻找一个计数为0的线性地址空间。计数为0说明该地址没有被映射
	*/
	for (;;) {
/* last_pkmap_nr为static,如果累加到头了,就在从头开始查询 */
		last_pkmap_nr = (last_pkmap_nr + 1) & LAST_PKMAP_MASK;
		if (!last_pkmap_nr) {
			flush_all_zero_pkmaps();
			count = LAST_PKMAP;
		}
		/* 找到了,返回 */
		if (!pkmap_count[last_pkmap_nr])
			break;	/* Found a usable entry */
		/*到头了,重置,不用再sleep了*/
		if (--count)
			continue;

		/*
		 * Sleep for somebody else to unmap their entries
		 */
		/*
	线性地址的永久映射的数目是有限的,如果找不到,只能sleep,
	等待别人释放后再用
		*/	
		{
			DECLARE_WAITQUEUE(wait, current);

			__set_current_state(TASK_UNINTERRUPTIBLE);
			add_wait_queue(&pkmap_map_wait, &wait);
			spin_unlock(&kmap_lock);
			schedule();
			remove_wait_queue(&pkmap_map_wait, &wait);
			spin_lock(&kmap_lock);

			/* Somebody else might have mapped it while we slept */
			if (page_address(page))
				return (unsigned long)page_address(page);

			/* Re-start */
			goto start;
		}
	}
	/*如果有可用的线性地址,根据数组号反向查询对应的线性地址*/
	vaddr = PKMAP_ADDR(last_pkmap_nr);
	set_pte(&(pkmap_page_table[last_pkmap_nr]), mk_pte(page, kmap_prot));
	/* 更新计数器 */
	pkmap_count[last_pkmap_nr] = 1;
	/* 将线性地址和对应的页描述符存到hash表 */
	set_page_address(page, (void *)vaddr);

	return vaddr;
}

/*
映射一个高端内存,传入为描述符地址,返回为线性地址
*/
void fastcall *kmap_high(struct page *page)
{
	unsigned long vaddr;

	/*
	 * For highmem pages, we can't trust "virtual" until
	 * after we have the lock.
	 *
	 * We cannot call this from interrupts, as it may block
	 */
	spin_lock(&kmap_lock);
	/* 如果这个内存已经被映射,那么直接返回即可 */
	vaddr = (unsigned long)page_address(page);
	if (!vaddr)/* 之前没有映射,就映射 */
		vaddr = map_new_virtual(page);

	/*计数
	根据ULK P308,必须远大于1才正确
	*/
	pkmap_count[PKMAP_NR(vaddr)]++;
	if (pkmap_count[PKMAP_NR(vaddr)] < 2)
		BUG();
	spin_unlock(&kmap_lock);
	return (void*) vaddr;
}

EXPORT_SYMBOL(kmap_high);

/*
取消一个内存映射
*/
void fastcall kunmap_high(struct page *page)
{
	unsigned long vaddr;
	unsigned long nr;
	int need_wakeup;

	spin_lock(&kmap_lock);
	vaddr = (unsigned long)page_address(page);
	if (!vaddr)
		BUG();
	nr = PKMAP_NR(vaddr);

/*
注意这里并没有彻底释放一个内存映射,只是表明该内存映射已经无用了,
然后唤醒等待队列(如果有任务在等待),否则直接退出。
*/

	/*
	 * A count must never go down to zero
	 * without a TLB flush!
	 */
	need_wakeup = 0;
	switch (--pkmap_count[nr]) {
	case 0:
		BUG();
	case 1:
		/*
		 * Avoid an unnecessary wake_up() function call.
		 * The common case is pkmap_count[] == 1, but
		 * no waiters.
		 * The tasks queued in the wait-queue are guarded
		 * by both the lock in the wait-queue-head and by
		 * the kmap_lock.  As the kmap_lock is held here,
		 * no need for the wait-queue-head's lock.  Simply
		 * test if the queue is empty.
		 */
		need_wakeup = waitqueue_active(&pkmap_map_wait);
	}
	spin_unlock(&kmap_lock);

	/* do wake-up, if needed, race-free outside of the spin lock */
	if (need_wakeup)
		wake_up(&pkmap_map_wait);
}

EXPORT_SYMBOL(kunmap_high);

。。。。。。
。。。。。。
。。。。。。


#if defined(HASHED_PAGE_VIRTUAL)

#define PA_HASH_ORDER	7

/*
 * Describes one page->virtual association
 */

/*
高端地址映射时的结构体,存储在hash链表中,便于快速查找
*/
struct page_address_map {
	struct page *page;/* 页表指针 */
	void *virtual;        /* 线性地址 */
	struct list_head list;/* hash使用开链法存储,list为链表用*/
};

/*
 * page_address_map freelist, allocated from page_address_maps.
 */
static struct list_head page_address_pool;	/* freelist */
static spinlock_t pool_lock;			/* protects page_address_pool */

/*
 * Hash table bucket
 */

/*
定义一个hash表,用来存储不同的page指针(页描述符地址)对应的线性地址。
这里主要是管理高端内存的,因为高端内存无法直接映射一个线性地址,
必须通过内核映射才可以。
链表法存储hash元素:由于hash表的特性,不同的page可能会有相同的hash,因此hash表的结构体中
有一个锁,用来对这个结构体进行加锁,有一个链表,这个链表连接了
不同的page指针对应的page_address_map结构体。
*/
static struct page_address_slot {
	struct list_head lh;			/* List of page_address_maps */
	spinlock_t lock;			/* Protect this bucket's list */
} ____cacheline_aligned_in_smp page_address_htable[1<<PA_HASH_ORDER];

/*
工具函数,用来根据page指针获取对应的hash入口
*/
static struct page_address_slot *page_slot(struct page *page)
{
	return &page_address_htable[hash_ptr(page, PA_HASH_ORDER)];
}

/*根据page指针(页描述符地址)获取对应的线性地址*/
void *page_address(struct page *page)
{
	unsigned long flags;
	void *ret;
	struct page_address_slot *pas;

/*
如果不是高端内存,是(NORMAL DMA),那么可以直接获取
*/
	if (!PageHighMem(page))
		return lowmem_page_address(page);

/*
如果是高端内存,就要查询映射表了
*/

/* 首先根据page指针查询对应的hash头 */
	pas = page_slot(page);
	ret = NULL;
/* 锁住对应的hash头 */
	spin_lock_irqsave(&pas->lock, flags);

/*如果hash链表不为空,说明有元素在该hash中,那么就进行遍历,
看下是否有对应的page
*/
	if (!list_empty(&pas->lh)) {
		struct page_address_map *pam;/* 链表链的其实是page_address_map指针 */

		list_for_each_entry(pam, &pas->lh, list) {
			if (pam->page == page) {/* 如果遍历到了,就返回对应的线性地址 */
				ret = pam->virtual;
				goto done;
			}
		}
	}
done:
	spin_unlock_irqrestore(&pas->lock, flags);
	return ret;
}

EXPORT_SYMBOL(page_address);

/*
将线性地址和对应的页描述符指针存到hash表中 
这里额外做了删除功能,如果线性地址为空,说明
从hash表中删除对应的条目
*/
void set_page_address(struct page *page, void *virtual)
{
	unsigned long flags;
	struct page_address_slot *pas;
	struct page_address_map *pam;

	BUG_ON(!PageHighMem(page));

/*
hash表中的page_address_map结构体并不是临时申请的,而是一开始就申请好
然后连接在page_address_pool中的,使用时从里面取一个,用完了在释放回去
*/
	pas = page_slot(page);
	if (virtual) {		/* Add */
		BUG_ON(list_empty(&page_address_pool));

		spin_lock_irqsave(&pool_lock, flags);
		pam = list_entry(page_address_pool.next,
				struct page_address_map, list);
		list_del(&pam->list);
		spin_unlock_irqrestore(&pool_lock, flags);

		pam->page = page;
		pam->virtual = virtual;

		spin_lock_irqsave(&pas->lock, flags);
		list_add_tail(&pam->list, &pas->lh);
		spin_unlock_irqrestore(&pas->lock, flags);
	} else {		/* Remove */
		spin_lock_irqsave(&pas->lock, flags);
		list_for_each_entry(pam, &pas->lh, list) {
			if (pam->page == page) {
				list_del(&pam->list);
				spin_unlock_irqrestore(&pas->lock, flags);
				spin_lock_irqsave(&pool_lock, flags);
				list_add_tail(&pam->list, &page_address_pool);
				spin_unlock_irqrestore(&pool_lock, flags);
				goto done;
			}
		}
		spin_unlock_irqrestore(&pas->lock, flags);
	}
done:
	return;
}

static struct page_address_map page_address_maps[LAST_PKMAP];

void __init page_address_init(void)
{
	int i;

	INIT_LIST_HEAD(&page_address_pool);
	for (i = 0; i < ARRAY_SIZE(page_address_maps); i++)
		list_add(&page_address_maps[i].list, &page_address_pool);
	for (i = 0; i < ARRAY_SIZE(page_address_htable); i++) {
		INIT_LIST_HEAD(&page_address_htable[i].lh);
		spin_lock_init(&page_address_htable[i].lock);
	}
	spin_lock_init(&pool_lock);
}

#endif	/* defined(CONFIG_HIGHMEM) && !defined(WANT_PAGE_VIRTUAL) */

 

Leave a Reply

2 × 5 =

此站点使用Akismet来减少垃圾评论。了解我们如何处理您的评论数据