filemap.c

ARM 嵌入式 系统 设计与实例开发 实验教材 二源码

 * writepage function - and this has to be
 * worked around in the VM layer..
 *
 * We
 *  - mark the page dirty again (but do NOT
 *    add it back to the inode dirty list, as
 *    that would livelock in fdatasync)
 *  - activate the page so that the page stealer
 *    doesn't try to write it out over and over
 *    again.
 */
int fail_writepage(struct page *page)
{
	/* Only activate on memory-pressure, not fsync.. */
	if (PageLaunder(page)) {
		activate_page(page);
		SetPageReferenced(page);
	}

	/* Set the page dirty again, unlock */
	SetPageDirty(page);
	UnlockPage(page);
	return 0;
}

EXPORT_SYMBOL(fail_writepage);

/**
 *      filemap_fdatasync - walk the list of dirty pages of the given address space
 *     	and writepage() all of them.
 * 
 *      @mapping: address space structure to write
 *
 */
int filemap_fdatasync(struct address_space * mapping)
{
	int ret = 0;
	int (*writepage)(struct page *) = mapping->a_ops->writepage;

	spin_lock(&pagecache_lock);

        while (!list_empty(&mapping->dirty_pages)) {
		struct page *page = list_entry(mapping->dirty_pages.next, struct page, list);

		list_del(&page->list);
		list_add(&page->list, &mapping->locked_pages);

		if (!PageDirty(page))
			continue;

		page_cache_get(page);
		spin_unlock(&pagecache_lock);

		lock_page(page);

		if (PageDirty(page)) {
			int err;
			ClearPageDirty(page);
			err = writepage(page);
			if (err && !ret)
				ret = err;
		} else
			UnlockPage(page);

		page_cache_release(page);
		spin_lock(&pagecache_lock);
	}
	spin_unlock(&pagecache_lock);
	return ret;
}

/**
 *      filemap_fdatawait - walk the list of locked pages of the given address space
 *     	and wait for all of them.
 * 
 *      @mapping: address space structure to wait for
 *
 */
int filemap_fdatawait(struct address_space * mapping)
{
	int ret = 0;

	spin_lock(&pagecache_lock);

        while (!list_empty(&mapping->locked_pages)) {
		struct page *page = list_entry(mapping->locked_pages.next, struct page, list);

		list_del(&page->list);
		list_add(&page->list, &mapping->clean_pages);

		if (!PageLocked(page))
			continue;

		page_cache_get(page);
		spin_unlock(&pagecache_lock);

		___wait_on_page(page);
		if (PageError(page))
			ret = -EIO;

		page_cache_release(page);
		spin_lock(&pagecache_lock);
	}
	spin_unlock(&pagecache_lock);
	return ret;
}

/*
 * Add a page to the inode page cache.
 *
 * The caller must have locked the page and 
 * set all the page flags correctly..
 */
void add_to_page_cache_locked(struct page * page, struct address_space *mapping, unsigned long index)
{
	if (!PageLocked(page))
		BUG();

	page->index = index;
	page_cache_get(page);
	spin_lock(&pagecache_lock);
	add_page_to_inode_queue(mapping, page);
	add_page_to_hash_queue(page, page_hash(mapping, index));
	spin_unlock(&pagecache_lock);

	lru_cache_add(page);
}

/*
 * This adds a page to the page cache, starting out as locked,
 * owned by us, but unreferenced, not uptodate and with no errors.
 */
static inline void __add_to_page_cache(struct page * page,
	struct address_space *mapping, unsigned long offset,
	struct page **hash)
{
	unsigned long flags;

	flags = page->flags & ~(1 << PG_uptodate | 1 << PG_error | 1 << PG_dirty | 1 << PG_referenced | 1 << PG_arch_1 | 1 << PG_checked);
	page->flags = flags | (1 << PG_locked);
	page_cache_get(page);
	page->index = offset;
	add_page_to_inode_queue(mapping, page);
	add_page_to_hash_queue(page, hash);
}

void add_to_page_cache(struct page * page, struct address_space * mapping, unsigned long offset)
{
	spin_lock(&pagecache_lock);
	__add_to_page_cache(page, mapping, offset, page_hash(mapping, offset));
	spin_unlock(&pagecache_lock);
	lru_cache_add(page);
}

int add_to_page_cache_unique(struct page * page,
	struct address_space *mapping, unsigned long offset,
	struct page **hash)
{
	int err;
	struct page *alias;

	spin_lock(&pagecache_lock);
	alias = __find_page_nolock(mapping, offset, *hash);

	err = 1;
	if (!alias) {
		__add_to_page_cache(page,mapping,offset,hash);
		err = 0;
	}

	spin_unlock(&pagecache_lock);
	if (!err)
		lru_cache_add(page);
	return err;
}

/*
 * This adds the requested page to the page cache if it isn't already there,
 * and schedules an I/O to read in its contents from disk.
 */
static int FASTCALL(page_cache_read(struct file * file, unsigned long offset));
static int page_cache_read(struct file * file, unsigned long offset)
{
	struct address_space *mapping = file->f_dentry->d_inode->i_mapping;
	struct page **hash = page_hash(mapping, offset);
	struct page *page; 

	spin_lock(&pagecache_lock);
	page = __find_page_nolock(mapping, offset, *hash);
	spin_unlock(&pagecache_lock);
	if (page)
		return 0;

	page = page_cache_alloc(mapping);
	if (!page)
		return -ENOMEM;

	if (!add_to_page_cache_unique(page, mapping, offset, hash)) {
		int error = mapping->a_ops->readpage(file, page);
		page_cache_release(page);
		return error;
	}
	/*
	 * We arrive here in the unlikely event that someone 
	 * raced with us and added our page to the cache first.
	 */
	page_cache_release(page);
	return 0;
}

/*
 * Read in an entire cluster at once.  A cluster is usually a 64k-
 * aligned block that includes the page requested in "offset."
 */
static int FASTCALL(read_cluster_nonblocking(struct file * file, unsigned long offset,
					     unsigned long filesize));
static int read_cluster_nonblocking(struct file * file, unsigned long offset,
	unsigned long filesize)
{
	unsigned long pages = CLUSTER_PAGES;

	offset = CLUSTER_OFFSET(offset);
	while ((pages-- > 0) && (offset < filesize)) {
		int error = page_cache_read(file, offset);
		if (error < 0)
			return error;
		offset ++;
	}

	return 0;
}

/* 
 * Wait for a page to get 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..
 */
void ___wait_on_page(struct page *page)
{
	struct task_struct *tsk = current;
	DECLARE_WAITQUEUE(wait, tsk);

	add_wait_queue(&page->wait, &wait);
	do {
		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
		if (!PageLocked(page))
			break;
		sync_page(page);
		schedule();
	} while (PageLocked(page));
	tsk->state = TASK_RUNNING;
	remove_wait_queue(&page->wait, &wait);
}

void unlock_page(struct page *page)
{
	clear_bit(PG_launder, &(page)->flags);
	smp_mb__before_clear_bit();
	if (!test_and_clear_bit(PG_locked, &(page)->flags))
		BUG();
	smp_mb__after_clear_bit(); 
	if (waitqueue_active(&(page)->wait))
	wake_up(&(page)->wait);
}

/*
 * Get a lock on the page, assuming we need to sleep
 * to get it..
 */
static void __lock_page(struct page *page)
{
	struct task_struct *tsk = current;
	DECLARE_WAITQUEUE(wait, tsk);

	add_wait_queue_exclusive(&page->wait, &wait);
	for (;;) {
		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
		if (PageLocked(page)) {
			sync_page(page);
			schedule();
		}
		if (!TryLockPage(page))
			break;
	}
	tsk->state = TASK_RUNNING;
	remove_wait_queue(&page->wait, &wait);
}
	

/*
 * Get an exclusive lock on the page, optimistically
 * assuming it's not locked..
 */
void lock_page(struct page *page)
{
	if (TryLockPage(page))
		__lock_page(page);
}

/*
 * a rather lightweight function, finding and getting a reference to a
 * hashed page atomically.
 */
struct page * __find_get_page(struct address_space *mapping,
			      unsigned long offset, struct page **hash)
{
	struct page *page;

	/*
	 * We scan the hash list read-only. Addition to and removal from
	 * the hash-list needs a held write-lock.
	 */
	spin_lock(&pagecache_lock);
	page = __find_page_nolock(mapping, offset, *hash);
	if (page)
		page_cache_get(page);
	spin_unlock(&pagecache_lock);
	return page;
}

/*
 * Same as above, but trylock it instead of incrementing the count.
 */
struct page *find_trylock_page(struct address_space *mapping, unsigned long offset)
{
	struct page *page;
	struct page **hash = page_hash(mapping, offset);

	spin_lock(&pagecache_lock);
	page = __find_page_nolock(mapping, offset, *hash);
	if (page) {
		if (TryLockPage(page))
			page = NULL;
	}
	spin_unlock(&pagecache_lock);
	return page;
}

/*
 * Must be called with the pagecache lock held,
 * will return with it held (but it may be dropped
 * during blocking operations..
 */
static struct page * FASTCALL(__find_lock_page_helper(struct address_space *, unsigned long, struct page *));
static struct page * __find_lock_page_helper(struct address_space *mapping,
					unsigned long offset, struct page *hash)
{
	struct page *page;

	/*
	 * We scan the hash list read-only. Addition to and removal from
	 * the hash-list needs a held write-lock.
	 */
repeat:
	page = __find_page_nolock(mapping, offset, hash);
	if (page) {
		page_cache_get(page);
		if (TryLockPage(page)) {
			spin_unlock(&pagecache_lock);
			lock_page(page);
			spin_lock(&pagecache_lock);

			/* Has the page been re-allocated while we slept? */
			if (page->mapping != mapping || page->index != offset) {
				UnlockPage(page);
				page_cache_release(page);
				goto repeat;
			}
		}
	}
	return page;
}

/*
 * Same as the above, but lock the page too, verifying that
 * it's still valid once we own it.
 */
struct page * __find_lock_page (struct address_space *mapping,
				unsigned long offset, struct page **hash)
{
	struct page *page;

	spin_lock(&pagecache_lock);
	page = __find_lock_page_helper(mapping, offset, *hash);
	spin_unlock(&pagecache_lock);
	return page;
}

/*
 * Same as above, but create the page if required..
 */
struct page * find_or_create_page(struct address_space *mapping, unsigned long index, unsigned int gfp_mask)
{
	struct page *page;
	struct page **hash = page_hash(mapping, index);

	spin_lock(&pagecache_lock);
	page = __find_lock_page_helper(mapping, index, *hash);
	spin_unlock(&pagecache_lock);
	if (!page) {
		struct page *newpage = alloc_page(gfp_mask);
		if (newpage) {
			spin_lock(&pagecache_lock);
			page = __find_lock_page_helper(mapping, index, *hash);
			if (likely(!page)) {
				page = newpage;
				__add_to_page_cache(page, mapping, index, hash);
				newpage = NULL;
			}
			spin_unlock(&pagecache_lock);
			if (newpage == NULL)
				lru_cache_add(page);
			else 
				page_cache_release(newpage);
		}
	}
	return page;	
}

/*
 * Returns locked page at given index in given cache, creating it if needed.
 */
struct page *grab_cache_page(struct address_space *mapping, unsigned long index)
{
	return find_or_create_page(mapping, index, mapping->gfp_mask);
}


/*
 * Same as grab_cache_page, but do not wait if the page is unavailable.
 * This is intended for speculative data generators, where the data can
 * be regenerated if the page couldn't be grabbed.  This routine should
 * be safe to call while holding the lock for another page.
 */
struct page *grab_cache_page_nowait(struct address_space *mapping, unsigned long index)
{
	struct page *page, **hash;

	hash = page_hash(mapping, index);
	page = __find_get_page(mapping, index, hash);

	if ( page ) {
		if ( !TryLockPage(page) ) {
			/* Page found and locked */
			/* This test is overly paranoid, but what the heck... */
			if ( unlikely(page->mapping != mapping || page->index != index) ) {
				/* Someone reallocated this page under us. */
				UnlockPage(page);
				page_cache_release(page);
				return NULL;
			} else {
				return page;
			}
		} else {
			/* Page locked by someone else */
			page_cache_release(page);
			return NULL;
		}
	}

	page = page_cache_alloc(mapping);
	if ( unlikely(!page) )
		return NULL;	/* Failed to allocate a page */

	if ( unlikely(add_to_page_cache_unique(page, mapping, index, hash)) ) {
		/* Someone else grabbed the page already. */
		page_cache_release(page);
		return NULL;
	}

	return page;
}

#if 0
#define PROFILE_READAHEAD
#define DEBUG_READAHEAD
#endif

/*
 * Read-ahead profiling information
 * --------------------------------
 * Every PROFILE_MAXREADCOUNT, the following information is written 
 * to the syslog:
 *   Percentage of asynchronous read-ahead.
 *   Average of read-ahead fields context value.
 * If DEBUG_READAHEAD is defined, a snapshot of these fields is written 
 * to the syslog.
 */

#ifdef PROFILE_READAHEAD

#define PROFILE_MAXREADCOUNT 1000

static unsigned long total_reada;
static unsigned long total_async;
static unsigned long total_ramax;
static unsigned long total_ralen;
static unsigned long total_rawin;

static void profile_readahead(int async, struct file *filp)
{
	unsigned long flags;

	++total_reada;
	if (async)
		++total_async;

	total_ramax	+= filp->f_ramax;
	total_ralen	+= filp->f_ralen;
	total_rawin	+= filp->f_rawin;

	if (total_reada > PROFILE_MAXREADCOUNT) {
		save_flags(flags);