f2fs: change the current atomic write way

Current atomic write has three major issues like below.
 - keeps the updates in non-reclaimable memory space and they are even
   hard to be migrated, which is not good for contiguous memory
   allocation.
 - disk spaces used for atomic files cannot be garbage collected, so
   this makes it difficult for the filesystem to be defragmented.
 - If atomic write operations hit the threshold of either memory usage
   or garbage collection failure count, All the atomic write operations
   will fail immediately.

To resolve the issues, I will keep a COW inode internally for all the
updates to be flushed from memory, when we need to flush them out in a
situation like high memory pressure. These COW inodes will be tagged
as orphan inodes to be reclaimed in case of sudden power-cut or system
failure during atomic writes.

Signed-off-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

1 files changed, 19 insertions, 9 deletions

diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index 37bdda931e0c..c549acb52ac4 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -840,8 +840,8 @@ out:
 }
 
 static int __f2fs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
-					struct dentry *dentry, umode_t mode,
-					struct inode **whiteout)
+			  struct dentry *dentry, umode_t mode, bool is_whiteout,
+			  struct inode **new_inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
 	struct inode *inode;
@@ -855,7 +855,7 @@ static int __f2fs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
 	if (IS_ERR(inode))
 		return PTR_ERR(inode);
 
-	if (whiteout) {
+	if (is_whiteout) {
 		init_special_inode(inode, inode->i_mode, WHITEOUT_DEV);
 		inode->i_op = &f2fs_special_inode_operations;
 	} else {
@@ -880,21 +880,25 @@ static int __f2fs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
 	f2fs_add_orphan_inode(inode);
 	f2fs_alloc_nid_done(sbi, inode->i_ino);
 
-	if (whiteout) {
+	if (is_whiteout) {
 		f2fs_i_links_write(inode, false);
 
 		spin_lock(&inode->i_lock);
 		inode->i_state |= I_LINKABLE;
 		spin_unlock(&inode->i_lock);
-
-		*whiteout = inode;
 	} else {
-		d_tmpfile(dentry, inode);
+		if (dentry)
+			d_tmpfile(dentry, inode);
+		else
+			f2fs_i_links_write(inode, false);
 	}
 	/* link_count was changed by d_tmpfile as well. */
 	f2fs_unlock_op(sbi);
 	unlock_new_inode(inode);
 
+	if (new_inode)
+		*new_inode = inode;
+
 	f2fs_balance_fs(sbi, true);
 	return 0;
 
@@ -915,7 +919,7 @@ static int f2fs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
 	if (!f2fs_is_checkpoint_ready(sbi))
 		return -ENOSPC;
 
-	return __f2fs_tmpfile(mnt_userns, dir, dentry, mode, NULL);
+	return __f2fs_tmpfile(mnt_userns, dir, dentry, mode, false, NULL);
 }
 
 static int f2fs_create_whiteout(struct user_namespace *mnt_userns,
@@ -925,7 +929,13 @@ static int f2fs_create_whiteout(struct user_namespace *mnt_userns,
 		return -EIO;
 
 	return __f2fs_tmpfile(mnt_userns, dir, NULL,
-				S_IFCHR | WHITEOUT_MODE, whiteout);
+				S_IFCHR | WHITEOUT_MODE, true, whiteout);
+}
+
+int f2fs_get_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
+		     struct inode **new_inode)
+{
+	return __f2fs_tmpfile(mnt_userns, dir, NULL, S_IFREG, false, new_inode);
 }
 
 static int f2fs_rename(struct user_namespace *mnt_userns, struct inode *old_dir,