f2fs: introduce a new global lock scheme

In the previous version, f2fs uses global locks according to the usage types,
such as directory operations, block allocation, block write, and so on.

Reference the following lock types in f2fs.h.
enum lock_type {
	RENAME,		/* for renaming operations */
	DENTRY_OPS,	/* for directory operations */
	DATA_WRITE,	/* for data write */
	DATA_NEW,	/* for data allocation */
	DATA_TRUNC,	/* for data truncate */
	NODE_NEW,	/* for node allocation */
	NODE_TRUNC,	/* for node truncate */
	NODE_WRITE,	/* for node write */
	NR_LOCK_TYPE,
};

In that case, we lose the performance under the multi-threading environment,
since every types of operations must be conducted one at a time.

In order to address the problem, let's share the locks globally with a mutex
array regardless of any types.
So, let users grab a mutex and perform their jobs in parallel as much as
possbile.

For this, I propose a new global lock scheme as follows.

0. Data structure
 - f2fs_sb_info -> mutex_lock[NR_GLOBAL_LOCKS]
 - f2fs_sb_info -> node_write

1. mutex_lock_op(sbi)
 - try to get an avaiable lock from the array.
 - returns the index of the gottern lock variable.

2. mutex_unlock_op(sbi, index of the lock)
 - unlock the given index of the lock.

3. mutex_lock_all(sbi)
 - grab all the locks in the array before the checkpoint.

4. mutex_unlock_all(sbi)
 - release all the locks in the array after checkpoint.

5. block_operations()
 - call mutex_lock_all()
 - sync_dirty_dir_inodes()
 - grab node_write
 - sync_node_pages()

Note that,
 the pairs of mutex_lock_op()/mutex_unlock_op() and
 mutex_lock_all()/mutex_unlock_all() should be used together.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>

1 files changed, 38 insertions, 39 deletions

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index cf9ff5f76134..72a1b30f720a 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -260,6 +260,9 @@ struct page *get_lock_data_page(struct inode *inode, pgoff_t index)
 /*
  * Caller ensures that this data page is never allocated.
  * A new zero-filled data page is allocated in the page cache.
+ *
+ * Also, caller should grab and release a mutex by calling mutex_lock_op() and
+ * mutex_unlock_op().
  */
 struct page *get_new_data_page(struct inode *inode, pgoff_t index,
 						bool new_i_size)
@@ -479,10 +482,11 @@ static int f2fs_write_data_page(struct page *page,
 	const pgoff_t end_index = ((unsigned long long) i_size)
 							>> PAGE_CACHE_SHIFT;
 	unsigned offset;
+	bool need_balance_fs = false;
 	int err = 0;
 
 	if (page->index < end_index)
-		goto out;
+		goto write;
 
 	/*
 	 * If the offset is out-of-range of file size,
@@ -494,50 +498,46 @@ static int f2fs_write_data_page(struct page *page,
 			dec_page_count(sbi, F2FS_DIRTY_DENTS);
 			inode_dec_dirty_dents(inode);
 		}
-		goto unlock_out;
+		goto out;
 	}
 
 	zero_user_segment(page, offset, PAGE_CACHE_SIZE);
-out:
-	if (sbi->por_doing)
-		goto redirty_out;
-
-	if (wbc->for_reclaim && !S_ISDIR(inode->i_mode) && !is_cold_data(page))
+write:
+	if (sbi->por_doing) {
+		err = AOP_WRITEPAGE_ACTIVATE;
 		goto redirty_out;
+	}
 
-	mutex_lock_op(sbi, DATA_WRITE);
+	/* Dentry blocks are controlled by checkpoint */
 	if (S_ISDIR(inode->i_mode)) {
 		dec_page_count(sbi, F2FS_DIRTY_DENTS);
 		inode_dec_dirty_dents(inode);
+		err = do_write_data_page(page);
+	} else {
+		int ilock = mutex_lock_op(sbi);
+		err = do_write_data_page(page);
+		mutex_unlock_op(sbi, ilock);
+		need_balance_fs = true;
 	}
-	err = do_write_data_page(page);
-	if (err && err != -ENOENT) {
-		wbc->pages_skipped++;
-		set_page_dirty(page);
-	}
-	mutex_unlock_op(sbi, DATA_WRITE);
+	if (err == -ENOENT)
+		goto out;
+	else if (err)
+		goto redirty_out;
 
 	if (wbc->for_reclaim)
 		f2fs_submit_bio(sbi, DATA, true);
 
-	if (err == -ENOENT)
-		goto unlock_out;
-
 	clear_cold_data(page);
+out:
 	unlock_page(page);
-
-	if (!wbc->for_reclaim && !S_ISDIR(inode->i_mode))
+	if (need_balance_fs)
 		f2fs_balance_fs(sbi);
 	return 0;
 
-unlock_out:
-	unlock_page(page);
-	return (err == -ENOENT) ? 0 : err;
-
 redirty_out:
 	wbc->pages_skipped++;
 	set_page_dirty(page);
-	return AOP_WRITEPAGE_ACTIVATE;
+	return err;
 }
 
 #define MAX_DESIRED_PAGES_WP	4096
@@ -592,6 +592,7 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 	pgoff_t index = ((unsigned long long) pos) >> PAGE_CACHE_SHIFT;
 	struct dnode_of_data dn;
 	int err = 0;
+	int ilock;
 
 	/* for nobh_write_end */
 	*fsdata = NULL;
@@ -603,28 +604,21 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 		return -ENOMEM;
 	*pagep = page;
 
-	mutex_lock_op(sbi, DATA_NEW);
+	ilock = mutex_lock_op(sbi);
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = get_dnode_of_data(&dn, index, ALLOC_NODE);
-	if (err) {
-		mutex_unlock_op(sbi, DATA_NEW);
-		f2fs_put_page(page, 1);
-		return err;
-	}
+	if (err)
+		goto err;
 
-	if (dn.data_blkaddr == NULL_ADDR) {
+	if (dn.data_blkaddr == NULL_ADDR)
 		err = reserve_new_block(&dn);
-		if (err) {
-			f2fs_put_dnode(&dn);
-			mutex_unlock_op(sbi, DATA_NEW);
-			f2fs_put_page(page, 1);
-			return err;
-		}
-	}
+
 	f2fs_put_dnode(&dn);
+	if (err)
+		goto err;
 
-	mutex_unlock_op(sbi, DATA_NEW);
+	mutex_unlock_op(sbi, ilock);
 
 	if ((len == PAGE_CACHE_SIZE) || PageUptodate(page))
 		return 0;
@@ -654,6 +648,11 @@ out:
 	SetPageUptodate(page);
 	clear_cold_data(page);
 	return 0;
+
+err:
+	mutex_unlock_op(sbi, ilock);
+	f2fs_put_page(page, 1);
+	return err;
 }
 
 static ssize_t f2fs_direct_IO(int rw, struct kiocb *iocb,