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, 22 insertions, 24 deletions

diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 155b362dad63..07be88ddb9f8 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -34,19 +34,18 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	block_t old_blk_addr;
 	struct dnode_of_data dn;
-	int err;
+	int err, ilock;
 
 	f2fs_balance_fs(sbi);
 
 	sb_start_pagefault(inode->i_sb);
 
-	mutex_lock_op(sbi, DATA_NEW);
-
 	/* block allocation */
+	ilock = mutex_lock_op(sbi);
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = get_dnode_of_data(&dn, page->index, ALLOC_NODE);
 	if (err) {
-		mutex_unlock_op(sbi, DATA_NEW);
+		mutex_unlock_op(sbi, ilock);
 		goto out;
 	}
 
@@ -56,13 +55,12 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 		err = reserve_new_block(&dn);
 		if (err) {
 			f2fs_put_dnode(&dn);
-			mutex_unlock_op(sbi, DATA_NEW);
+			mutex_unlock_op(sbi, ilock);
 			goto out;
 		}
 	}
 	f2fs_put_dnode(&dn);
-
-	mutex_unlock_op(sbi, DATA_NEW);
+	mutex_unlock_op(sbi, ilock);
 
 	lock_page(page);
 	if (page->mapping != inode->i_mapping ||
@@ -223,20 +221,19 @@ static int truncate_blocks(struct inode *inode, u64 from)
 	unsigned int blocksize = inode->i_sb->s_blocksize;
 	struct dnode_of_data dn;
 	pgoff_t free_from;
-	int count = 0;
+	int count = 0, ilock = -1;
 	int err;
 
 	free_from = (pgoff_t)
 			((from + blocksize - 1) >> (sbi->log_blocksize));
 
-	mutex_lock_op(sbi, DATA_TRUNC);
-
+	ilock = mutex_lock_op(sbi);
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE);
 	if (err) {
 		if (err == -ENOENT)
 			goto free_next;
-		mutex_unlock_op(sbi, DATA_TRUNC);
+		mutex_unlock_op(sbi, ilock);
 		return err;
 	}
 
@@ -247,6 +244,7 @@ static int truncate_blocks(struct inode *inode, u64 from)
 
 	count -= dn.ofs_in_node;
 	BUG_ON(count < 0);
+
 	if (dn.ofs_in_node || IS_INODE(dn.node_page)) {
 		truncate_data_blocks_range(&dn, count);
 		free_from += count;
@@ -255,7 +253,7 @@ static int truncate_blocks(struct inode *inode, u64 from)
 	f2fs_put_dnode(&dn);
 free_next:
 	err = truncate_inode_blocks(inode, free_from);
-	mutex_unlock_op(sbi, DATA_TRUNC);
+	mutex_unlock_op(sbi, ilock);
 
 	/* lastly zero out the first data page */
 	truncate_partial_data_page(inode, from);
@@ -363,15 +361,16 @@ static void fill_zero(struct inode *inode, pgoff_t index,
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	struct page *page;
+	int ilock;
 
 	if (!len)
 		return;
 
 	f2fs_balance_fs(sbi);
 
-	mutex_lock_op(sbi, DATA_NEW);
+	ilock = mutex_lock_op(sbi);
 	page = get_new_data_page(inode, index, false);
-	mutex_unlock_op(sbi, DATA_NEW);
+	mutex_unlock_op(sbi, ilock);
 
 	if (!IS_ERR(page)) {
 		wait_on_page_writeback(page);
@@ -388,13 +387,10 @@ int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end)
 
 	for (index = pg_start; index < pg_end; index++) {
 		struct dnode_of_data dn;
-		struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 
-		mutex_lock_op(sbi, DATA_TRUNC);
 		set_new_dnode(&dn, inode, NULL, NULL, 0);
 		err = get_dnode_of_data(&dn, index, LOOKUP_NODE);
 		if (err) {
-			mutex_unlock_op(sbi, DATA_TRUNC);
 			if (err == -ENOENT)
 				continue;
 			return err;
@@ -403,7 +399,6 @@ int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end)
 		if (dn.data_blkaddr != NULL_ADDR)
 			truncate_data_blocks_range(&dn, 1);
 		f2fs_put_dnode(&dn);
-		mutex_unlock_op(sbi, DATA_TRUNC);
 	}
 	return 0;
 }
@@ -434,6 +429,7 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len, int mode)
 			struct address_space *mapping = inode->i_mapping;
 			loff_t blk_start, blk_end;
 			struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
+			int ilock;
 
 			f2fs_balance_fs(sbi);
 
@@ -441,7 +437,10 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len, int mode)
 			blk_end = pg_end << PAGE_CACHE_SHIFT;
 			truncate_inode_pages_range(mapping, blk_start,
 					blk_end - 1);
+
+			ilock = mutex_lock_op(sbi);
 			ret = truncate_hole(inode, pg_start, pg_end);
+			mutex_unlock_op(sbi, ilock);
 		}
 	}
 
@@ -475,13 +474,13 @@ static int expand_inode_data(struct inode *inode, loff_t offset,
 
 	for (index = pg_start; index <= pg_end; index++) {
 		struct dnode_of_data dn;
+		int ilock;
 
-		mutex_lock_op(sbi, DATA_NEW);
-
+		ilock = mutex_lock_op(sbi);
 		set_new_dnode(&dn, inode, NULL, NULL, 0);
 		ret = get_dnode_of_data(&dn, index, ALLOC_NODE);
 		if (ret) {
-			mutex_unlock_op(sbi, DATA_NEW);
+			mutex_unlock_op(sbi, ilock);
 			break;
 		}
 
@@ -489,13 +488,12 @@ static int expand_inode_data(struct inode *inode, loff_t offset,
 			ret = reserve_new_block(&dn);
 			if (ret) {
 				f2fs_put_dnode(&dn);
-				mutex_unlock_op(sbi, DATA_NEW);
+				mutex_unlock_op(sbi, ilock);
 				break;
 			}
 		}
 		f2fs_put_dnode(&dn);
-
-		mutex_unlock_op(sbi, DATA_NEW);
+		mutex_unlock_op(sbi, ilock);
 
 		if (pg_start == pg_end)
 			new_size = offset + len;