Skip to contentSkip to conversation

f2fs: early check broken symlink length in the encrypted case

PATCH In review Jaegeuk Kim 2 participants 9 messages Dec 28, 2015 – Dec 30, 2015

Subsystems (2)
lore
Share & tools
lore

Jaegeuk Kim

author maintainer
#0 lore raw

PATCH 1/3 f2fs: early check broken symlink length in the encrypted case

If link is broken, its len is zero, and we don't need to move forward.

---
 fs/f2fs/namei.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
fs/f2fs/namei.c +8 -2
index fb41c80..5cc4128 100644
@@ -931,7 +931,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 {
 	struct page *cpage = NULL;
 	char *caddr, *paddr = NULL;
-	struct f2fs_str cstr;
+	struct f2fs_str cstr = FSTR_INIT(NULL, 0);
 	struct f2fs_str pstr = FSTR_INIT(NULL, 0);
 	struct inode *inode = d_inode(dentry);
 	struct f2fs_encrypted_symlink_data *sd;
@@ -952,6 +952,12 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 	/* Symlink is encrypted */
 	sd = (struct f2fs_encrypted_symlink_data *)caddr;
 	cstr.len = le16_to_cpu(sd->len);
+
+	/* this is broken symlink case */
+	if (cstr.len == 0) {
+		res = -ENOENT;
+		goto errout;
+	}
 	cstr.name = kmalloc(cstr.len, GFP_NOFS);
 	if (!cstr.name) {
 		res = -ENOMEM;
@@ -960,7 +966,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 	memcpy(cstr.name, sd->encrypted_path, cstr.len);
 
 	/* this is broken symlink case */
-	if (cstr.name[0] == 0 && cstr.len == 0) {
+	if (cstr.name[0] == 0) {
 		res = -ENOENT;
 		goto errout;
 	}
Signature · 1 line
2.5.4 (Apple Git-61)
author maintainer
#1 lore raw

PATCH 2/3 f2fs: use i_size_read to get i_size

We need to use i_size_read() to get inode->i_size.

---
 fs/f2fs/data.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
fs/f2fs/data.c +4 -3
index f34f42a..8a89810 100644
@@ -1386,10 +1386,11 @@ skip_write:
 static void f2fs_write_failed(struct address_space *mapping, loff_t to)
 {
 	struct inode *inode = mapping->host;
+	loff_t i_size = i_size_read(inode);
 
-	if (to > inode->i_size) {
-		truncate_pagecache(inode, inode->i_size);
-		truncate_blocks(inode, inode->i_size, true);
+	if (to > i_size) {
+		truncate_pagecache(inode, i_size);
+		truncate_blocks(inode, i_size, true);
 	}
 }
 
Signature · 1 line
2.5.4 (Apple Git-61)
#2 lore raw

RE: [f2fs-dev] [PATCH 2/3] f2fs: use i_size_read to get i_size

Hi Jaegeuk,

Quoted 9 lines

-----Original Message-----
From: Jaegeuk Kim [mailto:[email protected]]
Sent: Tuesday, December 29, 2015 7:32 AM
To: [email protected]; [email protected]; [email protected]
Cc: Jaegeuk Kim
Subject: [f2fs-dev] [PATCH 2/3] f2fs: use i_size_read to get i_size

We need to use i_size_read() to get inode->i_size.

All callers of f2fs_write_failed should be protected by i_mutex, so no one can change i_size, do we really need to use i_size_read here?

Thanks,

Quoted 34 lines
---
 fs/f2fs/data.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
fs/f2fs/data.c +4 -4
index f34f42a..8a89810 100644
@@ -1386,10 +1386,11 @@ skip_write:
 static void f2fs_write_failed(struct address_space *mapping, loff_t to)
 {
 	struct inode *inode = mapping->host;
+	loff_t i_size = i_size_read(inode);
 
-	if (to > inode->i_size) {
-		truncate_pagecache(inode, inode->i_size);
-		truncate_blocks(inode, inode->i_size, true);
+	if (to > i_size) {
+		truncate_pagecache(inode, i_size);
+		truncate_blocks(inode, i_size, true);
 	}
 }
 
--

2.5.4 (Apple Git-61)

------------------------------------------------------------------------------
_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

author maintainer
#3 lore raw

Re: [f2fs-dev] [PATCH 2/3] f2fs: use i_size_read to get i_size

Hi Chao,

Quoted 10 lines · Chao Yu

Hi Jaegeuk,

Quoted 8 lines

-----Original Message-----
From: Jaegeuk Kim [mailto:[email protected]]
Sent: Tuesday, December 29, 2015 7:32 AM
To: [email protected]; [email protected]; [email protected]
Cc: Jaegeuk Kim
Subject: [f2fs-dev] [PATCH 2/3] f2fs: use i_size_read to get i_size

Quoted 4 lines
Quoted 1 line

We need to use i_size_read() to get inode->i_size.

All callers of f2fs_write_failed should be protected by i_mutex, so no one can change i_size, do we really need to use i_size_read here?

Just to follow convention and avoid any confusion later. We also use i_size_read/write consistently across whole areas.

Thanks,

Quoted 37 lines

Thanks,

Quoted 34 lines
---
 fs/f2fs/data.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
fs/f2fs/data.c +4 -4
index f34f42a..8a89810 100644
@@ -1386,10 +1386,11 @@ skip_write:
 static void f2fs_write_failed(struct address_space *mapping, loff_t to)
 {
 	struct inode *inode = mapping->host;
+	loff_t i_size = i_size_read(inode);
 
-	if (to > inode->i_size) {
-		truncate_pagecache(inode, inode->i_size);
-		truncate_blocks(inode, inode->i_size, true);
+	if (to > i_size) {
+		truncate_pagecache(inode, i_size);
+		truncate_blocks(inode, i_size, true);
 	}
 }
 
--

2.5.4 (Apple Git-61)

------------------------------------------------------------------------------
_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

author maintainer
#4 lore raw

PATCH 3/3 f2fs: load largest extent all the time

In reply to Jaegeuk Kim

Otherwise, we can get mismatched largest extent information.

One example is:
1. mount f2fs w/ extent_cache
2. make a small extent
3. umount
4. mount f2fs w/o extent_cache
5. update the largest extent
6. umount
7. mount f2fs w/ extent_cache
8. get the old extent made by #2

---
 fs/f2fs/extent_cache.c | 18 +++++++++++++-----
 fs/f2fs/f2fs.h         |  2 +-
 fs/f2fs/inode.c        |  3 ++-
 3 files changed, 16 insertions(+), 7 deletions(-)
fs/f2fs/extent_cache.c +13 -5
index e7b6548..23e7c82 100644
@@ -166,20 +166,27 @@ static void __drop_largest_extent(struct inode *inode,
 		largest->len = 0;
 }
 
-void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
+/* return true, if inode page is changed */
+bool f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct extent_tree *et;
 	struct extent_node *en;
 	struct extent_info ei;
 
-	if (!f2fs_may_extent_tree(inode))
-		return;
+	if (!f2fs_may_extent_tree(inode)) {
+		/* drop largest extent */
+		if (i_ext && i_ext->len) {
+			i_ext->len = 0;
+			return true;
+		}
+		return false;
+	}
 
 	et = __grab_extent_tree(inode);
 
-	if (!i_ext || le32_to_cpu(i_ext->len) < F2FS_MIN_EXTENT_LEN)
-		return;
+	if (!i_ext || !i_ext->len)
+		return false;
 
 	set_extent_info(&ei, le32_to_cpu(i_ext->fofs),
 		le32_to_cpu(i_ext->blk), le32_to_cpu(i_ext->len));
@@ -196,6 +203,7 @@ void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 	}
 out:
 	write_unlock(&et->lock);
+	return false;
 }
 
 static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
fs/f2fs/f2fs.h +1 -1
index e04b2be..a339508 100644
@@ -2083,7 +2083,7 @@ void f2fs_leave_shrinker(struct f2fs_sb_info *);
  * extent_cache.c
  */
 unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *, int);
-void f2fs_init_extent_tree(struct inode *, struct f2fs_extent *);
+bool f2fs_init_extent_tree(struct inode *, struct f2fs_extent *);
 unsigned int f2fs_destroy_extent_node(struct inode *);
 void f2fs_destroy_extent_tree(struct inode *);
 bool f2fs_lookup_extent_cache(struct inode *, pgoff_t, struct extent_info *);
fs/f2fs/inode.c +2 -1
index ec3fb32..e955008 100644
@@ -138,7 +138,8 @@ static int do_read_inode(struct inode *inode)
 	fi->i_pino = le32_to_cpu(ri->i_pino);
 	fi->i_dir_level = ri->i_dir_level;
 
-	f2fs_init_extent_tree(inode, &ri->i_ext);
+	if (f2fs_init_extent_tree(inode, &ri->i_ext))
+		set_page_dirty(node_page);
 
 	get_inline_info(fi, ri);
 
Signature · 1 line
2.5.4 (Apple Git-61)
#5 lore raw

RE: [f2fs-dev] [PATCH 3/3] f2fs: load largest extent all the time

Hi Jaegeuk,

Quoted 10 lines

-----Original Message-----
From: Jaegeuk Kim [mailto:[email protected]]
Sent: Tuesday, December 29, 2015 7:32 AM
To: [email protected]; [email protected]; [email protected]
Cc: Jaegeuk Kim
Subject: [f2fs-dev] [PATCH 3/3] f2fs: load largest extent all the time

Otherwise, we can get mismatched largest extent information.

Quoted 9 lines

One example is:
1. mount f2fs w/ extent_cache
2. make a small extent
3. umount
4. mount f2fs w/o extent_cache
5. update the largest extent
6. umount
7. mount f2fs w/ extent_cache
8. get the old extent made by #2

Good catch!

It's a pity to drop it since it may provide some help if we can reuse it.

Thanks,

Quoted 89 lines
---
 fs/f2fs/extent_cache.c | 18 +++++++++++++-----
 fs/f2fs/f2fs.h         |  2 +-
 fs/f2fs/inode.c        |  3 ++-
 3 files changed, 16 insertions(+), 7 deletions(-)
fs/f2fs/extent_cache.c +13 -5
index e7b6548..23e7c82 100644
@@ -166,20 +166,27 @@ static void __drop_largest_extent(struct inode *inode,
 		largest->len = 0;
 }
 
-void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
+/* return true, if inode page is changed */
+bool f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct extent_tree *et;
 	struct extent_node *en;
 	struct extent_info ei;
 
-	if (!f2fs_may_extent_tree(inode))
-		return;
+	if (!f2fs_may_extent_tree(inode)) {
+		/* drop largest extent */
+		if (i_ext && i_ext->len) {
+			i_ext->len = 0;
+			return true;
+		}
+		return false;
+	}
 
 	et = __grab_extent_tree(inode);
 
-	if (!i_ext || le32_to_cpu(i_ext->len) < F2FS_MIN_EXTENT_LEN)
-		return;
+	if (!i_ext || !i_ext->len)
+		return false;
 
 	set_extent_info(&ei, le32_to_cpu(i_ext->fofs),
 		le32_to_cpu(i_ext->blk), le32_to_cpu(i_ext->len));
@@ -196,6 +203,7 @@ void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 	}
 out:
 	write_unlock(&et->lock);
+	return false;
 }
 
 static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
fs/f2fs/f2fs.h +1 -1
index e04b2be..a339508 100644
@@ -2083,7 +2083,7 @@ void f2fs_leave_shrinker(struct f2fs_sb_info *);
  * extent_cache.c
  */
 unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *, int);
-void f2fs_init_extent_tree(struct inode *, struct f2fs_extent *);
+bool f2fs_init_extent_tree(struct inode *, struct f2fs_extent *);
 unsigned int f2fs_destroy_extent_node(struct inode *);
 void f2fs_destroy_extent_tree(struct inode *);
 bool f2fs_lookup_extent_cache(struct inode *, pgoff_t, struct extent_info *);
fs/f2fs/inode.c +2 -2
index ec3fb32..e955008 100644
@@ -138,7 +138,8 @@ static int do_read_inode(struct inode *inode)
 	fi->i_pino = le32_to_cpu(ri->i_pino);
 	fi->i_dir_level = ri->i_dir_level;
 
-	f2fs_init_extent_tree(inode, &ri->i_ext);
+	if (f2fs_init_extent_tree(inode, &ri->i_ext))
+		set_page_dirty(node_page);
 
 	get_inline_info(fi, ri);
 
--

2.5.4 (Apple Git-61)

------------------------------------------------------------------------------
_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

author maintainer
#6 lore raw

Re: [f2fs-dev] [PATCH 3/3] f2fs: load largest extent all the time

Hi Chao,

Quoted 22 lines · Chao Yu

Hi Jaegeuk,

Quoted 10 lines

-----Original Message-----
From: Jaegeuk Kim [mailto:[email protected]]
Sent: Tuesday, December 29, 2015 7:32 AM
To: [email protected]; [email protected]; [email protected]
Cc: Jaegeuk Kim
Subject: [f2fs-dev] [PATCH 3/3] f2fs: load largest extent all the time

Otherwise, we can get mismatched largest extent information.

Quoted 9 lines

One example is:
1. mount f2fs w/ extent_cache
2. make a small extent
3. umount
4. mount f2fs w/o extent_cache
5. update the largest extent
6. umount
7. mount f2fs w/ extent_cache
8. get the old extent made by #2

Quoted 4 lines

Good catch!

It's a pity to drop it since it may provide some help if we can reuse it.

Indeed, but I expect it's not a common case.

Thanks,

Quoted 92 lines

Thanks,

Quoted 89 lines
---
 fs/f2fs/extent_cache.c | 18 +++++++++++++-----
 fs/f2fs/f2fs.h         |  2 +-
 fs/f2fs/inode.c        |  3 ++-
 3 files changed, 16 insertions(+), 7 deletions(-)
fs/f2fs/extent_cache.c +13 -5
index e7b6548..23e7c82 100644
@@ -166,20 +166,27 @@ static void __drop_largest_extent(struct inode *inode,
 		largest->len = 0;
 }
 
-void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
+/* return true, if inode page is changed */
+bool f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct extent_tree *et;
 	struct extent_node *en;
 	struct extent_info ei;
 
-	if (!f2fs_may_extent_tree(inode))
-		return;
+	if (!f2fs_may_extent_tree(inode)) {
+		/* drop largest extent */
+		if (i_ext && i_ext->len) {
+			i_ext->len = 0;
+			return true;
+		}
+		return false;
+	}
 
 	et = __grab_extent_tree(inode);
 
-	if (!i_ext || le32_to_cpu(i_ext->len) < F2FS_MIN_EXTENT_LEN)
-		return;
+	if (!i_ext || !i_ext->len)
+		return false;
 
 	set_extent_info(&ei, le32_to_cpu(i_ext->fofs),
 		le32_to_cpu(i_ext->blk), le32_to_cpu(i_ext->len));
@@ -196,6 +203,7 @@ void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 	}
 out:
 	write_unlock(&et->lock);
+	return false;
 }
 
 static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
fs/f2fs/f2fs.h +1 -1
index e04b2be..a339508 100644
@@ -2083,7 +2083,7 @@ void f2fs_leave_shrinker(struct f2fs_sb_info *);
  * extent_cache.c
  */
 unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *, int);
-void f2fs_init_extent_tree(struct inode *, struct f2fs_extent *);
+bool f2fs_init_extent_tree(struct inode *, struct f2fs_extent *);
 unsigned int f2fs_destroy_extent_node(struct inode *);
 void f2fs_destroy_extent_tree(struct inode *);
 bool f2fs_lookup_extent_cache(struct inode *, pgoff_t, struct extent_info *);
fs/f2fs/inode.c +2 -2
index ec3fb32..e955008 100644
@@ -138,7 +138,8 @@ static int do_read_inode(struct inode *inode)
 	fi->i_pino = le32_to_cpu(ri->i_pino);
 	fi->i_dir_level = ri->i_dir_level;
 
-	f2fs_init_extent_tree(inode, &ri->i_ext);
+	if (f2fs_init_extent_tree(inode, &ri->i_ext))
+		set_page_dirty(node_page);
 
 	get_inline_info(fi, ri);
 
--

2.5.4 (Apple Git-61)

------------------------------------------------------------------------------
_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

#7 lore raw
In reply to Jaegeuk Kim

Hi Jaegeuk,

Quoted 30 lines

-----Original Message-----
From: Jaegeuk Kim [mailto:[email protected]]
Sent: Tuesday, December 29, 2015 7:32 AM
To: [email protected]; [email protected]; [email protected]
Cc: Jaegeuk Kim
Subject: [f2fs-dev] [PATCH 1/3] f2fs: early check broken symlink length in the encrypted case

If link is broken, its len is zero, and we don't need to move forward.

---
 fs/f2fs/namei.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
fs/f2fs/namei.c +1 -1
index fb41c80..5cc4128 100644
@@ -931,7 +931,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void
 **cook
 {
 	struct page *cpage = NULL;
 	char *caddr, *paddr = NULL;
-	struct f2fs_str cstr;
+	struct f2fs_str cstr = FSTR_INIT(NULL, 0);
 	struct f2fs_str pstr = FSTR_INIT(NULL, 0);
 	struct inode *inode = d_inode(dentry);
fs/f2fs/namei.c +0 -0
 	struct f2fs_encrypted_symlink_data *sd;

fs/f2fs/namei.c static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void · line 958

Quoted 7 lines
fs/f2fs/namei.c +3 -0
@@ -952,6 +952,12 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void
 **cook
 	/* Symlink is encrypted */
 	sd = (struct f2fs_encrypted_symlink_data *)caddr;
 	cstr.len = le16_to_cpu(sd->len);
+
+	/* this is broken symlink case */
+	if (cstr.len == 0) {

This case seems rare, can we add unlikely here?

fs/f2fs/namei.c static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void · line 970

Quoted 13 lines
fs/f2fs/namei.c +3 -0
+		res = -ENOENT;
+		goto errout;
+	}
 	cstr.name = kmalloc(cstr.len, GFP_NOFS);
 	if (!cstr.name) {
fs/f2fs/namei.c +1 -1
 		res = -ENOMEM;
@@ -960,7 +966,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void
 **cook
 	memcpy(cstr.name, sd->encrypted_path, cstr.len);
 
 	/* this is broken symlink case */
-	if (cstr.name[0] == 0 && cstr.len == 0) {
+	if (cstr.name[0] == 0) {

ditto.

Thanks,

Quoted 12 lines
fs/f2fs/namei.c +0 -0
 		res = -ENOENT;
 		goto errout;

}
--
2.5.4 (Apple Git-61)

------------------------------------------------------------------------------
_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

author maintainer
#8 lore raw

Hi Chao,

Quoted 33 lines · Chao Yu

Hi Jaegeuk,

Quoted 31 lines

-----Original Message-----
From: Jaegeuk Kim [mailto:[email protected]]
Sent: Tuesday, December 29, 2015 7:32 AM
To: [email protected]; [email protected]; [email protected]
Cc: Jaegeuk Kim
Subject: [f2fs-dev] [PATCH 1/3] f2fs: early check broken symlink length in the encrypted case

If link is broken, its len is zero, and we don't need to move forward.

---
 fs/f2fs/namei.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
fs/f2fs/namei.c +1 -1
index fb41c80..5cc4128 100644
@@ -931,7 +931,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void
 **cook
 {
 	struct page *cpage = NULL;
 	char *caddr, *paddr = NULL;
-	struct f2fs_str cstr;
+	struct f2fs_str cstr = FSTR_INIT(NULL, 0);
 	struct f2fs_str pstr = FSTR_INIT(NULL, 0);
 	struct inode *inode = d_inode(dentry);
fs/f2fs/namei.c +0 -0
 	struct f2fs_encrypted_symlink_data *sd;
@@ -952,6 +952,12 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void
 **cook
Quoted 8 lines

fs/f2fs/namei.c static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook · line 957

Quoted 6 lines
fs/f2fs/namei.c +3 -0
 	/* Symlink is encrypted */
 	sd = (struct f2fs_encrypted_symlink_data *)caddr;
 	cstr.len = le16_to_cpu(sd->len);
+
+	/* this is broken symlink case */
+	if (cstr.len == 0) {

This case seems rare, can we add unlikely here?

Agreed.
Will change them.

Thanks,

Quoted 31 lines
Quoted 8 lines
fs/f2fs/namei.c +3 -0
+		res = -ENOENT;
+		goto errout;
+	}
 	cstr.name = kmalloc(cstr.len, GFP_NOFS);
 	if (!cstr.name) {
fs/f2fs/namei.c +0 -0
 		res = -ENOMEM;

fs/f2fs/namei.c static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void · line 970

Quoted 6 lines
fs/f2fs/namei.c +1 -1
@@ -960,7 +966,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void
 **cook
 	memcpy(cstr.name, sd->encrypted_path, cstr.len);
 
 	/* this is broken symlink case */
-	if (cstr.name[0] == 0 && cstr.len == 0) {
+	if (cstr.name[0] == 0) {

ditto.

Thanks,

Quoted 12 lines
fs/f2fs/namei.c +0 -0
 		res = -ENOENT;
 		goto errout;

}
--
2.5.4 (Apple Git-61)

------------------------------------------------------------------------------
_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel