[f2fs-dev] [PATCH 1/3] f2fs: iput only if whole data blocks are flushed

Subsystems: f2fs file system, filesystems (vfs and infrastructure), the rest

3 messages, 1 author, 2013-05-28 · open the first message on its own page

[f2fs-dev] [PATCH 1/3] f2fs: iput only if whole data blocks are flushed

From: Jaegeuk Kim <hidden>
Date: 2013-05-28 01:19:43

If there remains some unwritten blocks from the recovery, we should not call
iput on that directory inode.
Otherwise, we can loose some dentry blocks after the recovery.

Signed-off-by: Jaegeuk Kim <redacted>
---
 fs/f2fs/checkpoint.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 01ddc91..0d3701d 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c

@@ -501,8 +501,10 @@ void remove_dirty_dir_inode(struct inode *inode)
 		return;
 
 	spin_lock(&sbi->dir_inode_lock);
-	if (atomic_read(&F2FS_I(inode)->dirty_dents))
-		goto out;
+	if (atomic_read(&F2FS_I(inode)->dirty_dents)) {
+		spin_unlock(&sbi->dir_inode_lock);
+		return;
+	}
 
 	list_for_each(this, head) {
 		struct dir_inode_entry *entry;

@@ -516,7 +518,6 @@ void remove_dirty_dir_inode(struct inode *inode)
 			break;
 		}
 	}
-out:
 	spin_unlock(&sbi->dir_inode_lock);
 
 	/* Only from the recovery routine */

-- 
1.8.1.3.566.gaa39828


------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may

[f2fs-dev] [PATCH 2/3] f2fs: fix dentry recovery routine

From: Jaegeuk Kim <hidden>
Date: 2013-05-28 01:19:44

The error scenario is:
1. create /a
(1.a link /a /b)
2. sync
3. unlinke /a
4. create /a
5. fsync /a
6. Sudden power-off

When the f2fs recovers the fsynced dentry, /a, we discover an exsiting dentry at
f2fs_find_entry() in recover_dentry().

In such the case, we should unlink the existing dentry and its inode
and then recover newly created dentry.

Signed-off-by: Jaegeuk Kim <redacted>
---
 fs/f2fs/recovery.c | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 0dd2ce1..85f5e54 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c

@@ -44,9 +44,10 @@ static int recover_dentry(struct page *ipage, struct inode *inode)
 	struct f2fs_node *raw_node = (struct f2fs_node *)kaddr;
 	struct f2fs_inode *raw_inode = &(raw_node->i);
 	nid_t pino = le32_to_cpu(raw_inode->i_pino);
+	struct f2fs_dir_entry *de;
 	struct qstr name;
 	struct page *page;
-	struct inode *dir;
+	struct inode *dir, *einode;
 	int err = 0;
 
 	dir = check_dirty_dir_inode(F2FS_SB(inode->i_sb), pino);

@@ -61,13 +62,21 @@ static int recover_dentry(struct page *ipage, struct inode *inode)
 
 	name.len = le32_to_cpu(raw_inode->i_namelen);
 	name.name = raw_inode->i_name;
-
-	if (f2fs_find_entry(dir, &name, &page)) {
-		kunmap(page);
-		f2fs_put_page(page, 0);
-	} else {
-		err = __f2fs_add_link(dir, &name, inode);
+retry:
+	de = f2fs_find_entry(dir, &name, &page);
+	if (de) {
+		einode = f2fs_iget(inode->i_sb, le32_to_cpu(de->ino));
+		if (IS_ERR(einode)) {
+			WARN_ON(1);
+			if (PTR_ERR(einode) == -ENOENT)
+				err = -EEXIST;
+			goto out;
+		}
+		f2fs_delete_entry(de, page, einode);
+		iput(einode);
+		goto retry;
 	}
+	err = __f2fs_add_link(dir, &name, inode);
 out:
 	f2fs_msg(inode->i_sb, KERN_NOTICE, "recover_inode and its dentry: "
 			"ino = %x, name = %s, dir = %lx, err = %d",

-- 
1.8.1.3.566.gaa39828


------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may

[f2fs-dev] [PATCH 3/3] f2fs: fix incorrect iputs during the dentry recovery

From: Jaegeuk Kim <hidden>
Date: 2013-05-28 01:19:45

- iget/iput flow in the dentry recovery process

1. *dir* = f2fs_iget
2. set FI_DELAY_IPUT to *dir*
3. add *dir* to the dirty_dir_list
		   - __f2fs_add_link
		     - recover_dentry)
4. iput *dir* by remove_dirty_dir_inode
		   - sync_dirty_dir_inodes
		     - write_chekcpoint

If *dir*'s i_count is not 1 (i.e., root dir), remove_dirty_dir_inode is called
later and then iput is triggered again due to the FI_DELAY_IPUT flag.
So, let's unset the flag properly once iput is triggered.

Signed-off-by: Jaegeuk Kim <redacted>
---
 fs/f2fs/checkpoint.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 0d3701d..6f56e57 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c

@@ -521,8 +521,10 @@ void remove_dirty_dir_inode(struct inode *inode)
 	spin_unlock(&sbi->dir_inode_lock);
 
 	/* Only from the recovery routine */
-	if (is_inode_flag_set(F2FS_I(inode), FI_DELAY_IPUT))
+	if (is_inode_flag_set(F2FS_I(inode), FI_DELAY_IPUT)) {
+		clear_inode_flag(F2FS_I(inode), FI_DELAY_IPUT);
 		iput(inode);
+	}
 }
 
 struct inode *check_dirty_dir_inode(struct f2fs_sb_info *sbi, nid_t ino)

-- 
1.8.1.3.566.gaa39828


------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help