Skip to contentSkip to conversation

ext4: use fsdata to track inline data write state and fix race

PATCH Applied Aditya Srivastava 3 participants 5 messages Jul 3 – Jul 23

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v4 1/2 ext4: use fsdata to track inline data write state and fix race

From: Aditya Prakash Srivastava <[email protected]>

Instead of checking the live inode state (ext4_has_inline_data(inode) and ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) in the write_end handlers, use the fsdata parameter of the address space operations to explicitly pass down the state in which write_begin prepared the write.

A concurrent thread (such as ext4_page_mkwrite()) can convert the inline data to an extent between write_begin and write_end. If this happens, the write_end handlers would previously miss the inline write_end path and fall through to extent-based write_end logic. However, since block buffers were never allocated in write_begin, this resulted in NULL pointer dereferences or data loss because folio_buffers(folio) was NULL.

Define EXT4_WRITE_DATA_INLINE (4) as a bit flag (Bit 2), treating fsdata as bitwise flags rather than mutually exclusive enums to keep states of the write path independent. Communicate this state via fsdata:
1) ext4_write_begin() and ext4_da_write_begin() set the EXT4_WRITE_DATA_INLINE bit in *fsdata via bitwise OR when an inline write is successfully prepared.
2) On entry, ext4_write_begin() clears the EXT4_WRITE_DATA_INLINE bit to safely handle VFS retries (where generic_perform_write() bypasses the fsdata initialization on its retry jump).
3) The write_end handlers perform a bitwise AND to check if the EXT4_WRITE_DATA_INLINE bit is set and invoke the inline write_end helper accordingly.

Furthermore, during a buffered write, ext4_write_inline_data_end() acquires the xattr lock after preparing the write. If a concurrent page fault (ext4_page_mkwrite()) converts the inline data to an extent after the write_end handlers check the state but before ext4_write_inline_data_end() acquires the xattr write lock, the subsequent check will trigger a kernel panic via BUG_ON(!ext4_has_inline_data(inode)).

To keep git history working and bisectability clean, replace the BUG_ON check in ext4_write_inline_data_end() with a graceful error- handling retry path in this same commit. If the inline data is cleared after locking the xattr, we safely release all resources (releasing iloc.bh, unlocking/putting the folio, stopping the active journal transaction handle) and return 0 (VFS retry) to let the generic write path retry the operation safely.

Reported-by [email protected]
Fixes 3fdcfb668fd7 ("ext4: add journalled write support for inline data")

---
Changes in v4:
- Resent the bug fix (Patch 1) and the cleanup (Patch 2) as a unified 2-patch series to ensure clean upstream application and perfect logical separation, as suggested by Jan Kara.

Changes in v3:
- Changed EXT4_WRITE_DATA_INLINE to 4 (Bit 2) and treat fsdata as bitwise flags, allowing decoupling of independent states in the write path (e.g. standard fallback and inline writes), as suggested by Jan Kara.
- Clear the EXT4_WRITE_DATA_INLINE bit on entry to ext4_write_begin() to safely unroll state for any VFS write retries.
- Perform bitwise AND checks for EXT4_WRITE_DATA_INLINE and FALL_BACK_TO_NONDELALLOC flags in the write_end and da_write_end handlers.

Changes in v2:
- Folded the BUG_ON fix from the second patch into the first one to ensure bisectability across git history, as suggested by Jan Kara.
- Removed the pointless initialization *fsdata = NULL on entry to ext4_write_begin().
- Removed the redundant check if (fsdata) in ext4_write_begin().

 fs/ext4/ext4.h   |  1 +
 fs/ext4/inline.c | 14 +++++++++++++-
 fs/ext4/inode.c  | 24 +++++++++++++-----------
 3 files changed, 27 insertions(+), 12 deletions(-)
fs/ext4/ext4.h +1 -0
index b37c136ea3ab..9e6f6467bdeb 100644
@@ -3138,6 +3138,7 @@ int do_journal_get_write_access(handle_t *handle, struct inode *inode,
 void ext4_set_inode_mapping_order(struct inode *inode);
 #define FALL_BACK_TO_NONDELALLOC 1
 #define CONVERT_INLINE_DATA	 2
+#define EXT4_WRITE_DATA_INLINE	 4
 
 typedef enum {
 	EXT4_IGET_NORMAL =	0,
fs/ext4/inline.c +13 -1
index 8045e4ff270c..cfd591dc1d9c 100644
@@ -812,7 +812,19 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
 			goto out;
 		}
 		ext4_write_lock_xattr(inode, &no_expand);
-		BUG_ON(!ext4_has_inline_data(inode));
+		/*
+		 * We could have raced with ext4_page_mkwrite() converting
+		 * the inode and clearing the inline data flag, so we just
+		 * release resources and retry the whole write.
+		 */
+		if (unlikely(!ext4_has_inline_data(inode))) {
+			ext4_write_unlock_xattr(inode, &no_expand);
+			brelse(iloc.bh);
+			folio_unlock(folio);
+			folio_put(folio);
+			ext4_journal_stop(handle);
+			return 0;
+		}
 
 		/*
 		 * ei->i_inline_off may have changed since
fs/ext4/inode.c +13 -11
index ce99807c5f5b..4e1bf54e511d 100644
@@ -1302,6 +1302,8 @@ static int ext4_write_begin(const struct kiocb *iocb,
 	if (unlikely(ret))
 		return ret;
 
+	*fsdata = (void *)((unsigned long)*fsdata & ~EXT4_WRITE_DATA_INLINE);
+
 	trace_ext4_write_begin(inode, pos, len);
 	/*
 	 * Reserve one block more for addition to orphan list in case
@@ -1316,8 +1318,10 @@ static int ext4_write_begin(const struct kiocb *iocb,
 						    foliop);
 		if (ret < 0)
 			return ret;
-		if (ret == 1)
+		if (ret == 1) {
+			*fsdata = (void *)((unsigned long)*fsdata | EXT4_WRITE_DATA_INLINE);
 			return 0;
+		}
 	}
 
 	/*
@@ -1450,8 +1454,7 @@ static int ext4_write_end(const struct kiocb *iocb,
 
 	trace_ext4_write_end(inode, pos, len, copied);
 
-	if (ext4_has_inline_data(inode) &&
-	    ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA))
+	if ((unsigned long)fsdata & EXT4_WRITE_DATA_INLINE)
 		return ext4_write_inline_data_end(inode, pos, len, copied,
 						  folio);
 
@@ -1560,8 +1563,7 @@ static int ext4_journalled_write_end(const struct kiocb *iocb,
 
 	BUG_ON(!ext4_handle_valid(handle));
 
-	if (ext4_has_inline_data(inode) &&
-	    ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA))
+	if ((unsigned long)fsdata & EXT4_WRITE_DATA_INLINE)
 		return ext4_write_inline_data_end(inode, pos, len, copied,
 						  folio);
 
@@ -3161,8 +3163,10 @@ static int ext4_da_write_begin(const struct kiocb *iocb,
 						     foliop, fsdata, true);
 		if (ret < 0)
 			return ret;
-		if (ret == 1)
+		if (ret == 1) {
+			*fsdata = (void *)((unsigned long)*fsdata | EXT4_WRITE_DATA_INLINE);
 			return 0;
+		}
 	}
 
 retry:
@@ -3291,17 +3295,15 @@ static int ext4_da_write_end(const struct kiocb *iocb,
 			     struct folio *folio, void *fsdata)
 {
 	struct inode *inode = mapping->host;
-	int write_mode = (int)(unsigned long)fsdata;
+	unsigned long write_mode = (unsigned long)fsdata;
 
-	if (write_mode == FALL_BACK_TO_NONDELALLOC)
+	if (write_mode & FALL_BACK_TO_NONDELALLOC)
 		return ext4_write_end(iocb, mapping, pos,
 				      len, copied, folio, fsdata);
 
 	trace_ext4_da_write_end(inode, pos, len, copied);
 
-	if (write_mode != CONVERT_INLINE_DATA &&
-	    ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) &&
-	    ext4_has_inline_data(inode))
+	if (write_mode & EXT4_WRITE_DATA_INLINE)
 		return ext4_write_inline_data_end(inode, pos, len, copied,
 						  folio);
 
Signature · 1 line
2.47.3
#1 lore raw

PATCH v4 2/2 ext4: cleanup unused CONVERT_INLINE_DATA flag

From: Aditya Prakash Srivastava <[email protected]>

After implementing bitwise flags for tracking the inline data write state in the address space fsdata parameter, the CONVERT_INLINE_DATA state flag is left unused and can be removed.

Perform this clean-up by:
1) Deleting the CONVERT_INLINE_DATA definition from ext4.h.
2) Removing the void **fsdata argument from both the forward declaration and the definition of the internal helper ext4_da_convert_inline_data_to_extent().
3) Removing the void **fsdata argument from the declaration and definition of ext4_generic_write_inline_data() and updating the caller ext4_try_to_write_inline_data() and the internal re-alloc retry logic accordingly.
4) Updating ext4_da_write_begin() to call ext4_generic_write_inline_data() without the fsdata parameter.

---
Changes in v4:
- Resent as Patch 2/2 of the v4 series, as suggested by Jan Kara.

Changes in v3:
- Initial version of the cleanup patch.

 fs/ext4/ext4.h   |  5 ++---
 fs/ext4/inline.c | 13 +++++--------
 fs/ext4/inode.c  |  2 +-
 3 files changed, 8 insertions(+), 12 deletions(-)
fs/ext4/ext4.h +2 -3
index 9e6f6467bdeb..c2e4262406a7 100644
@@ -3137,8 +3137,7 @@ int do_journal_get_write_access(handle_t *handle, struct inode *inode,
 				struct buffer_head *bh);
 void ext4_set_inode_mapping_order(struct inode *inode);
 #define FALL_BACK_TO_NONDELALLOC 1
-#define CONVERT_INLINE_DATA	 2
-#define EXT4_WRITE_DATA_INLINE	 4
+#define EXT4_WRITE_DATA_INLINE	 2
 
 typedef enum {
 	EXT4_IGET_NORMAL =	0,
@@ -3749,7 +3748,7 @@ extern int ext4_generic_write_inline_data(struct address_space *mapping,
 					  struct inode *inode,
 					  loff_t pos, unsigned len,
 					  struct folio **foliop,
-					  void **fsdata, bool da);
+					  bool da);
 extern int ext4_try_add_inline_entry(handle_t *handle,
 				     struct ext4_filename *fname,
 				     struct inode *dir, struct inode *inode);
fs/ext4/inline.c +5 -8
index cfd591dc1d9c..0cb2dc195d23 100644
@@ -22,8 +22,7 @@
 
 
 static int ext4_da_convert_inline_data_to_extent(struct address_space *mapping,
-						 struct inode *inode,
-						 void **fsdata);
+						 struct inode *inode);
 
 static int ext4_get_inline_size(struct inode *inode)
 {
@@ -697,7 +696,7 @@ int ext4_generic_write_inline_data(struct address_space *mapping,
 					  struct inode *inode,
 					  loff_t pos, unsigned len,
 					  struct folio **foliop,
-					  void **fsdata, bool da)
+					  bool da)
 {
 	int ret;
 	handle_t *handle;
@@ -728,7 +727,7 @@ int ext4_generic_write_inline_data(struct address_space *mapping,
 			return ext4_convert_inline_data_to_extent(mapping, inode);
 		}
 
-		ret = ext4_da_convert_inline_data_to_extent(mapping, inode, fsdata);
+		ret = ext4_da_convert_inline_data_to_extent(mapping, inode);
 		if (ret == -ENOSPC &&
 		    ext4_should_retry_alloc(inode->i_sb, &retries))
 			goto retry_journal;
@@ -788,7 +787,7 @@ int ext4_try_to_write_inline_data(struct address_space *mapping,
 	if (pos + len > ext4_get_max_inline_size(inode))
 		return ext4_convert_inline_data_to_extent(mapping, inode);
 	return ext4_generic_write_inline_data(mapping, inode, pos, len,
-					      foliop, NULL, false);
+					      foliop, false);
 }
 
 int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
@@ -895,8 +894,7 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
  *    need to start the journal since the file's metadata isn't changed now.
  */
 static int ext4_da_convert_inline_data_to_extent(struct address_space *mapping,
-						 struct inode *inode,
-						 void **fsdata)
+						 struct inode *inode)
 {
 	int ret = 0, inline_size;
 	struct folio *folio;
@@ -934,7 +932,6 @@ static int ext4_da_convert_inline_data_to_extent(struct address_space *mapping,
 	folio_mark_dirty(folio);
 	folio_mark_uptodate(folio);
 	ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
-	*fsdata = (void *)CONVERT_INLINE_DATA;
 
 out:
 	up_read(&EXT4_I(inode)->xattr_sem);
fs/ext4/inode.c +1 -1
index 4e1bf54e511d..9a0fbaf73ce2 100644
@@ -3160,7 +3160,7 @@ static int ext4_da_write_begin(const struct kiocb *iocb,
 
 	if (ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) {
 		ret = ext4_generic_write_inline_data(mapping, inode, pos, len,
-						     foliop, fsdata, true);
+						     foliop, true);
 		if (ret < 0)
 			return ret;
 		if (ret == 1) {
Signature · 1 line
2.47.3
reviewer Reviewed-by on v4 2/2
#2 lore raw

Re: [PATCH v4 2/2] ext4: cleanup unused CONVERT_INLINE_DATA flag

Quoted 18 lines · Aditya Srivastava

From: Aditya Prakash Srivastava <[email protected]>

After implementing bitwise flags for tracking the inline data write state in the address space fsdata parameter, the CONVERT_INLINE_DATA state flag is left unused and can be removed.

Perform this clean-up by:
1) Deleting the CONVERT_INLINE_DATA definition from ext4.h.
2) Removing the void **fsdata argument from both the forward declaration and the definition of the internal helper ext4_da_convert_inline_data_to_extent().
3) Removing the void **fsdata argument from the declaration and definition of ext4_generic_write_inline_data() and updating the caller ext4_try_to_write_inline_data() and the internal re-alloc retry logic accordingly.
4) Updating ext4_da_write_begin() to call ext4_generic_write_inline_data() without the fsdata parameter.

Quoted 2 lines

Looks good! Feel free to add:

Honza

Quoted 110 lines

---
Changes in v4:
- Resent as Patch 2/2 of the v4 series, as suggested by Jan Kara.

Changes in v3:
- Initial version of the cleanup patch.

 fs/ext4/ext4.h   |  5 ++---
 fs/ext4/inline.c | 13 +++++--------
 fs/ext4/inode.c  |  2 +-
 3 files changed, 8 insertions(+), 12 deletions(-)
fs/ext4/ext4.h +2 -3
index 9e6f6467bdeb..c2e4262406a7 100644
@@ -3137,8 +3137,7 @@ int do_journal_get_write_access(handle_t *handle, struct inode *inode,
 				struct buffer_head *bh);
 void ext4_set_inode_mapping_order(struct inode *inode);
 #define FALL_BACK_TO_NONDELALLOC 1
-#define CONVERT_INLINE_DATA	 2
-#define EXT4_WRITE_DATA_INLINE	 4
+#define EXT4_WRITE_DATA_INLINE	 2
 
 typedef enum {
 	EXT4_IGET_NORMAL =	0,
@@ -3749,7 +3748,7 @@ extern int ext4_generic_write_inline_data(struct address_space *mapping,
 					  struct inode *inode,
 					  loff_t pos, unsigned len,
 					  struct folio **foliop,
-					  void **fsdata, bool da);
+					  bool da);
 extern int ext4_try_add_inline_entry(handle_t *handle,
 				     struct ext4_filename *fname,
 				     struct inode *dir, struct inode *inode);
fs/ext4/inline.c +5 -8
index cfd591dc1d9c..0cb2dc195d23 100644
@@ -22,8 +22,7 @@
 
 
 static int ext4_da_convert_inline_data_to_extent(struct address_space *mapping,
-						 struct inode *inode,
-						 void **fsdata);
+						 struct inode *inode);
 
 static int ext4_get_inline_size(struct inode *inode)
 {
@@ -697,7 +696,7 @@ int ext4_generic_write_inline_data(struct address_space *mapping,
 					  struct inode *inode,
 					  loff_t pos, unsigned len,
 					  struct folio **foliop,
-					  void **fsdata, bool da)
+					  bool da)
 {
 	int ret;
 	handle_t *handle;
@@ -728,7 +727,7 @@ int ext4_generic_write_inline_data(struct address_space *mapping,
 			return ext4_convert_inline_data_to_extent(mapping, inode);
 		}
 
-		ret = ext4_da_convert_inline_data_to_extent(mapping, inode, fsdata);
+		ret = ext4_da_convert_inline_data_to_extent(mapping, inode);
 		if (ret == -ENOSPC &&
 		    ext4_should_retry_alloc(inode->i_sb, &retries))
 			goto retry_journal;
@@ -788,7 +787,7 @@ int ext4_try_to_write_inline_data(struct address_space *mapping,
 	if (pos + len > ext4_get_max_inline_size(inode))
 		return ext4_convert_inline_data_to_extent(mapping, inode);
 	return ext4_generic_write_inline_data(mapping, inode, pos, len,
-					      foliop, NULL, false);
+					      foliop, false);
 }
 
 int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
@@ -895,8 +894,7 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
  *    need to start the journal since the file's metadata isn't changed now.
  */
 static int ext4_da_convert_inline_data_to_extent(struct address_space *mapping,
-						 struct inode *inode,
-						 void **fsdata)
+						 struct inode *inode)
 {
 	int ret = 0, inline_size;
 	struct folio *folio;
@@ -934,7 +932,6 @@ static int ext4_da_convert_inline_data_to_extent(struct address_space *mapping,
 	folio_mark_dirty(folio);
 	folio_mark_uptodate(folio);
 	ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
-	*fsdata = (void *)CONVERT_INLINE_DATA;
 
 out:
 	up_read(&EXT4_I(inode)->xattr_sem);
fs/ext4/inode.c +1 -1
index 4e1bf54e511d..9a0fbaf73ce2 100644
@@ -3160,7 +3160,7 @@ static int ext4_da_write_begin(const struct kiocb *iocb,
 
 	if (ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) {
 		ret = ext4_generic_write_inline_data(mapping, inode, pos, len,
-						     foliop, fsdata, true);
+						     foliop, true);
 		if (ret < 0)
 			return ret;
 		if (ret == 1) {
Signature · 1 line
2.47.3
Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR
reviewer Reviewed-by on v4 1/2
#3 lore raw
In reply to Aditya Srivastava
Quoted 46 lines · Aditya Srivastava

From: Aditya Prakash Srivastava <[email protected]>

Instead of checking the live inode state (ext4_has_inline_data(inode) and ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) in the write_end handlers, use the fsdata parameter of the address space operations to explicitly pass down the state in which write_begin prepared the write.

A concurrent thread (such as ext4_page_mkwrite()) can convert the inline data to an extent between write_begin and write_end. If this happens, the write_end handlers would previously miss the inline write_end path and fall through to extent-based write_end logic. However, since block buffers were never allocated in write_begin, this resulted in NULL pointer dereferences or data loss because folio_buffers(folio) was NULL.

Define EXT4_WRITE_DATA_INLINE (4) as a bit flag (Bit 2), treating fsdata as bitwise flags rather than mutually exclusive enums to keep states of the write path independent. Communicate this state via fsdata:
1) ext4_write_begin() and ext4_da_write_begin() set the EXT4_WRITE_DATA_INLINE bit in *fsdata via bitwise OR when an inline write is successfully prepared.
2) On entry, ext4_write_begin() clears the EXT4_WRITE_DATA_INLINE bit to safely handle VFS retries (where generic_perform_write() bypasses the fsdata initialization on its retry jump).
3) The write_end handlers perform a bitwise AND to check if the EXT4_WRITE_DATA_INLINE bit is set and invoke the inline write_end helper accordingly.

Furthermore, during a buffered write, ext4_write_inline_data_end() acquires the xattr lock after preparing the write. If a concurrent page fault (ext4_page_mkwrite()) converts the inline data to an extent after the write_end handlers check the state but before ext4_write_inline_data_end() acquires the xattr write lock, the subsequent check will trigger a kernel panic via BUG_ON(!ext4_has_inline_data(inode)).

To keep git history working and bisectability clean, replace the BUG_ON check in ext4_write_inline_data_end() with a graceful error- handling retry path in this same commit. If the inline data is cleared after locking the xattr, we safely release all resources (releasing iloc.bh, unlocking/putting the folio, stopping the active journal transaction handle) and return 0 (VFS retry) to let the generic write path retry the operation safely.

Quoted 5 lines
Reported-by [email protected]
Fixes 3fdcfb668fd7 ("ext4: add journalled write support for inline data")

Looks good! Feel free to add:

Honza

Quoted 147 lines

---
Changes in v4:
- Resent the bug fix (Patch 1) and the cleanup (Patch 2) as a unified 2-patch series to ensure clean upstream application and perfect logical separation, as suggested by Jan Kara.

Changes in v3:
- Changed EXT4_WRITE_DATA_INLINE to 4 (Bit 2) and treat fsdata as bitwise flags, allowing decoupling of independent states in the write path (e.g. standard fallback and inline writes), as suggested by Jan Kara.
- Clear the EXT4_WRITE_DATA_INLINE bit on entry to ext4_write_begin() to safely unroll state for any VFS write retries.
- Perform bitwise AND checks for EXT4_WRITE_DATA_INLINE and FALL_BACK_TO_NONDELALLOC flags in the write_end and da_write_end handlers.

Changes in v2:
- Folded the BUG_ON fix from the second patch into the first one to ensure bisectability across git history, as suggested by Jan Kara.
- Removed the pointless initialization *fsdata = NULL on entry to ext4_write_begin().
- Removed the redundant check if (fsdata) in ext4_write_begin().

 fs/ext4/ext4.h   |  1 +
 fs/ext4/inline.c | 14 +++++++++++++-
 fs/ext4/inode.c  | 24 +++++++++++++-----------
 3 files changed, 27 insertions(+), 12 deletions(-)
fs/ext4/ext4.h +1 -0
index b37c136ea3ab..9e6f6467bdeb 100644
@@ -3138,6 +3138,7 @@ int do_journal_get_write_access(handle_t *handle, struct inode *inode,
 void ext4_set_inode_mapping_order(struct inode *inode);
 #define FALL_BACK_TO_NONDELALLOC 1
 #define CONVERT_INLINE_DATA	 2
+#define EXT4_WRITE_DATA_INLINE	 4
 
 typedef enum {
 	EXT4_IGET_NORMAL =	0,
fs/ext4/inline.c +13 -1
index 8045e4ff270c..cfd591dc1d9c 100644
@@ -812,7 +812,19 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
 			goto out;
 		}
 		ext4_write_lock_xattr(inode, &no_expand);
-		BUG_ON(!ext4_has_inline_data(inode));
+		/*
+		 * We could have raced with ext4_page_mkwrite() converting
+		 * the inode and clearing the inline data flag, so we just
+		 * release resources and retry the whole write.
+		 */
+		if (unlikely(!ext4_has_inline_data(inode))) {
+			ext4_write_unlock_xattr(inode, &no_expand);
+			brelse(iloc.bh);
+			folio_unlock(folio);
+			folio_put(folio);
+			ext4_journal_stop(handle);
+			return 0;
+		}
 
 		/*
 		 * ei->i_inline_off may have changed since
fs/ext4/inode.c +13 -11
index ce99807c5f5b..4e1bf54e511d 100644
@@ -1302,6 +1302,8 @@ static int ext4_write_begin(const struct kiocb *iocb,
 	if (unlikely(ret))
 		return ret;
 
+	*fsdata = (void *)((unsigned long)*fsdata & ~EXT4_WRITE_DATA_INLINE);
+
 	trace_ext4_write_begin(inode, pos, len);
 	/*
 	 * Reserve one block more for addition to orphan list in case
@@ -1316,8 +1318,10 @@ static int ext4_write_begin(const struct kiocb *iocb,
 						    foliop);
 		if (ret < 0)
 			return ret;
-		if (ret == 1)
+		if (ret == 1) {
+			*fsdata = (void *)((unsigned long)*fsdata | EXT4_WRITE_DATA_INLINE);
 			return 0;
+		}
 	}
 
 	/*
@@ -1450,8 +1454,7 @@ static int ext4_write_end(const struct kiocb *iocb,
 
 	trace_ext4_write_end(inode, pos, len, copied);
 
-	if (ext4_has_inline_data(inode) &&
-	    ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA))
+	if ((unsigned long)fsdata & EXT4_WRITE_DATA_INLINE)
 		return ext4_write_inline_data_end(inode, pos, len, copied,
 						  folio);
 
@@ -1560,8 +1563,7 @@ static int ext4_journalled_write_end(const struct kiocb *iocb,
 
 	BUG_ON(!ext4_handle_valid(handle));
 
-	if (ext4_has_inline_data(inode) &&
-	    ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA))
+	if ((unsigned long)fsdata & EXT4_WRITE_DATA_INLINE)
 		return ext4_write_inline_data_end(inode, pos, len, copied,
 						  folio);
 
@@ -3161,8 +3163,10 @@ static int ext4_da_write_begin(const struct kiocb *iocb,
 						     foliop, fsdata, true);
 		if (ret < 0)
 			return ret;
-		if (ret == 1)
+		if (ret == 1) {
+			*fsdata = (void *)((unsigned long)*fsdata | EXT4_WRITE_DATA_INLINE);
 			return 0;
+		}
 	}
 
 retry:
@@ -3291,17 +3295,15 @@ static int ext4_da_write_end(const struct kiocb *iocb,
 			     struct folio *folio, void *fsdata)
 {
 	struct inode *inode = mapping->host;
-	int write_mode = (int)(unsigned long)fsdata;
+	unsigned long write_mode = (unsigned long)fsdata;
 
-	if (write_mode == FALL_BACK_TO_NONDELALLOC)
+	if (write_mode & FALL_BACK_TO_NONDELALLOC)
 		return ext4_write_end(iocb, mapping, pos,
 				      len, copied, folio, fsdata);
 
 	trace_ext4_da_write_end(inode, pos, len, copied);
 
-	if (write_mode != CONVERT_INLINE_DATA &&
-	    ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) &&
-	    ext4_has_inline_data(inode))
+	if (write_mode & EXT4_WRITE_DATA_INLINE)
 		return ext4_write_inline_data_end(inode, pos, len, copied,
 						  folio);
 
Signature · 1 line
2.47.3
Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR
maintainer Applied
#4 lore raw
In reply to Aditya Srivastava
Quoted 14 lines · Aditya Srivastava

Instead of checking the live inode state (ext4_has_inline_data(inode) and ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) in the write_end handlers, use the fsdata parameter of the address space operations to explicitly pass down the state in which write_begin prepared the write.

A concurrent thread (such as ext4_page_mkwrite()) can convert the inline data to an extent between write_begin and write_end. If this happens, the write_end handlers would previously miss the inline write_end path and fall through to extent-based write_end logic. However, since block buffers were never allocated in write_begin, this resulted in NULL pointer dereferences or data loss because folio_buffers(folio) was NULL.

Quoted 1 line

[...]

Applied, thanks!

[1/2] ext4: use fsdata to track inline data write state and fix race

      commit: 7edbb323bab2b2a609016014caafdb651c898249
[2/2] ext4: cleanup unused CONVERT_INLINE_DATA flag
      commit: f6065852d2137734bcb26c9f94e5daf526179aa9

Best regards,

Signature · 1 line
Theodore Ts'o <[email protected]>