[PATCH 04/15] f2fs: fix BUG_ON during f2fs_evict_inode(dir)

[Jaegeuk Kim]

During the dentry recovery routine, recover_inode() triggers __f2fs_add_link
with its directory inode.

In the following scenario, a bug is captured.
 1. dir = f2fs_iget(pino)
 2. __f2fs_add_link(dir, name)
 3. iput(dir)
  -> f2fs_evict_inode() faces with BUG_ON(atomic_read(fi->dirty_dents))

Kernel BUG at ffffffffa01c0676 [verbose debug info unavailable]
[<ffffffffa01c0676>] f2fs_evict_inode+0x276/0x300 [f2fs]
Call Trace:
 [<ffffffff8118ea00>] evict+0xb0/0x1b0
 [<ffffffff8118f1c5>] iput+0x105/0x190
 [<ffffffffa01d2dac>] recover_fsync_data+0x3bc/0x1070 [f2fs]
 [<ffffffff81692e8a>] ? io_schedule+0xaa/0xd0
 [<ffffffff81690acb>] ? __wait_on_bit_lock+0x7b/0xc0
 [<ffffffff8111a0e7>] ? __lock_page+0x67/0x70
 [<ffffffff81165e21>] ? kmem_cache_alloc+0x31/0x140
 [<ffffffff8118a502>] ? __d_instantiate+0x92/0xf0
 [<ffffffff812a949b>] ? security_d_instantiate+0x1b/0x30
 [<ffffffff8118a5b4>] ? d_instantiate+0x54/0x70

This means that we should flush all the dentry pages between iget and iput().
But, during the recovery routine, it is unallowed due to consistency, so we
have to wait the whole recovery process.
And then, write_checkpoint flushes all the dirty dentry blocks, and nicely we
can put the stale dir inodes from the dirty_dir_inode_list.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@xxxxxxxxxxx>
---
 fs/f2fs/checkpoint.c | 23 +++++++++++++++++++++++
 fs/f2fs/f2fs.h       |  2 ++
 fs/f2fs/recovery.c   | 14 +++++++++-----
 3 files changed, 34 insertions(+), 5 deletions(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index b1de01d..3d11449 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -514,6 +514,29 @@ void remove_dirty_dir_inode(struct inode *inode)
 	}
 out:
 	spin_unlock(&sbi->dir_inode_lock);
+
+	/* Only from the recovery routine */
+	if (is_inode_flag_set(F2FS_I(inode), FI_DELAY_IPUT))
+		iput(inode);
+}
+
+struct inode *check_dirty_dir_inode(struct f2fs_sb_info *sbi, nid_t ino)
+{
+	struct list_head *head = &sbi->dir_inode_list;
+	struct list_head *this;
+	struct inode *inode = NULL;
+
+	spin_lock(&sbi->dir_inode_lock);
+	list_for_each(this, head) {
+		struct dir_inode_entry *entry;
+		entry = list_entry(this, struct dir_inode_entry, list);
+		if (entry->inode->i_ino == ino) {
+			inode = entry->inode;
+			break;
+		}
+	}
+	spin_unlock(&sbi->dir_inode_lock);
+	return inode;
 }
 
 void sync_dirty_dir_inodes(struct f2fs_sb_info *sbi)
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 20aab02..ef6cac8 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -846,6 +846,7 @@ enum {
 	FI_INC_LINK,		/* need to increment i_nlink */
 	FI_ACL_MODE,		/* indicate acl mode */
 	FI_NO_ALLOC,		/* should not allocate any blocks */
+	FI_DELAY_IPUT,		/* used for the recovery */
 };
 
 static inline void set_inode_flag(struct f2fs_inode_info *fi, int flag)
@@ -1012,6 +1013,7 @@ int recover_orphan_inodes(struct f2fs_sb_info *);
 int get_valid_checkpoint(struct f2fs_sb_info *);
 void set_dirty_dir_page(struct inode *, struct page *);
 void remove_dirty_dir_inode(struct inode *);
+struct inode *check_dirty_dir_inode(struct f2fs_sb_info *, nid_t);
 void sync_dirty_dir_inodes(struct f2fs_sb_info *);
 void write_checkpoint(struct f2fs_sb_info *, bool);
 void init_orphan_info(struct f2fs_sb_info *);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index f77aedd..c573944 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -42,6 +42,7 @@ static int recover_dentry(struct page *ipage, struct inode *inode)
 {
 	struct f2fs_node *raw_node = (struct f2fs_node *)kmap(ipage);
 	struct f2fs_inode *raw_inode = &(raw_node->i);
+	nid_t pino = le32_to_cpu(raw_inode->i_pino);
 	struct qstr name;
 	struct f2fs_dir_entry *de;
 	struct page *page;
@@ -51,10 +52,14 @@ static int recover_dentry(struct page *ipage, struct inode *inode)
 	if (!is_dent_dnode(ipage))
 		goto out;
 
-	dir = f2fs_iget(inode->i_sb, le32_to_cpu(raw_inode->i_pino));
-	if (IS_ERR(dir)) {
-		err = PTR_ERR(dir);
-		goto out;
+	dir = check_dirty_dir_inode(F2FS_SB(inode->i_sb), pino);
+	if (!dir) {
+		dir = f2fs_iget(inode->i_sb, pino);
+		if (IS_ERR(dir)) {
+			err = PTR_ERR(dir);
+			goto out;
+		}
+		set_inode_flag(F2FS_I(dir), FI_DELAY_IPUT);
 	}
 
 	name.len = le32_to_cpu(raw_inode->i_namelen);
@@ -67,7 +72,6 @@ static int recover_dentry(struct page *ipage, struct inode *inode)
 	} else {
 		err = __f2fs_add_link(dir, &name, inode);
 	}
-	iput(dir);
 out:
 	kunmap(ipage);
 	return err;
-- 
1.8.1.3.566.gaa39828

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/