Skip to contentSkip to conversation

debugfs: disallow NULL string creation and fix callers

PATCH New Gui-Dong Han 1 participant 4 messages Mar 23

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v2 cover debugfs: disallow NULL string creation and fix callers

A recent bug report [1] highlighted a NULL pointer dereference when reading a debugfs string file created with a NULL pointer. The community discussed the issue and agreed that creating string nodes with NULL is invalid and should be forbidden at creation time [2]. Since no fix was submitted following the discussion, I have implemented the agreed solution.

Patch 1 modifies debugfs_create_str() to reject NULL pointers, returning early and triggering a WARN_ON to expose offending callers.

Patch 2 is a code hygiene fix. While modifying the file, I noticed the EXPORT_SYMBOL_GPL for debugfs_create_str() was misplaced far away from the function body. This patch moves it to the correct location.

I carefully audited existing callers across the kernel tree. Some drivers passing NULL have already been independently identified and fixed [3]. The remaining subsystem (soundwire) is addressed in patch 3 by initializing its string parameter to an empty string (""). The existing code correctly and safely handles empty strings.

While auditing, I also noticed a devm_kstrdup/kfree mismatch in the interconnect subsystem which has been addressed in a separate patch [4].

Changes in v2:
- Link to v1: https://lore.kernel.org/driver-core/[email protected]/
- Patch 1 & 2: No changes.
- Patch 3: Rewritten to use kstrdup() instead of devm_kstrdup() to match debugfs's kfree() behavior. Moved initialization to sdw_debugfs_init()
to correctly handle the global pointer and avoid overwriting during
slave probe.
- Patch 4 (drm/i915): Dropped. Further analysis confirmed those specific pointers are not passed to debugfs_create_str() [5].

[1] https://lore.kernel.org/lkml/[email protected]/
[2] https://lore.kernel.org/lkml/2025122221-gag-malt-75ba@gregkh/
[3] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=8cc27f5c6dd1
[4] https://lore.kernel.org/linux-pm/[email protected]/
[5] https://lore.kernel.org/driver-core/[email protected]/

Gui-Dong Han (3):
debugfs: check for NULL pointer in debugfs_create_str()
debugfs: fix placement of EXPORT_SYMBOL_GPL for debugfs_create_str()
soundwire: debugfs: initialize firmware_file to empty string

 drivers/soundwire/debugfs.c | 9 +++++++--
 fs/debugfs/file.c           | 7 +++++--
 2 files changed, 12 insertions(+), 4 deletions(-)
Signature · 1 line
2.43.0
#1 lore raw

PATCH v2 1/3 debugfs: check for NULL pointer in debugfs_create_str()

Passing a NULL pointer to debugfs_create_str() leads to a NULL pointer dereference when the debugfs file is read. Following upstream discussions, forbid the creation of debugfs string files with NULL pointers. Add a WARN_ON() to expose offending callers and return early.

---
v2:
* No changes.

---
 fs/debugfs/file.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
fs/debugfs/file.c +4 -1
index 3376ab6a519d..a941d73251b0 100644
@@ -1127,7 +1127,7 @@ static const struct file_operations fops_str_wo = {
  *          directory dentry if set.  If this parameter is %NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @value: a pointer to the variable that the file should read to and write
- *         from.
+ *         from. This pointer and the string it points to must not be %NULL.
  *
  * This function creates a file in debugfs with the given name that
  * contains the value of the variable @value.  If the @mode variable is so
@@ -1136,6 +1136,9 @@ static const struct file_operations fops_str_wo = {
 void debugfs_create_str(const char *name, umode_t mode,
 			struct dentry *parent, char **value)
 {
+	if (WARN_ON(!value || !*value))
+		return;
+
 	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_str,
 				   &fops_str_ro, &fops_str_wo);
 }
Signature · 1 line
2.43.0
#2 lore raw

PATCH v2 2/3 debugfs: fix placement of EXPORT_SYMBOL_GPL for debugfs_create_str()

In reply to Gui-Dong Han

The EXPORT_SYMBOL_GPL() for debugfs_create_str was placed incorrectly away from the function definition. Move it immediately below the debugfs_create_str() function where it belongs.

Fixes d60b59b96795 ("debugfs: Export debugfs_create_str symbol")

---
v2:
* No changes.

---
 fs/debugfs/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/debugfs/file.c +1 -1
index a941d73251b0..edd6aafbfbaa 100644
@@ -1047,7 +1047,6 @@ ssize_t debugfs_read_file_str(struct file *file, char __user *user_buf,
 
 	return ret;
 }
-EXPORT_SYMBOL_GPL(debugfs_create_str);
 
 static ssize_t debugfs_write_file_str(struct file *file, const char __user *user_buf,
 				      size_t count, loff_t *ppos)
@@ -1142,6 +1141,7 @@ void debugfs_create_str(const char *name, umode_t mode,
 	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_str,
 				   &fops_str_ro, &fops_str_wo);
 }
+EXPORT_SYMBOL_GPL(debugfs_create_str);
 
 static ssize_t read_file_blob(struct file *file, char __user *user_buf,
 			      size_t count, loff_t *ppos)
Signature · 1 line
2.43.0
#3 lore raw

PATCH v2 3/3 soundwire: debugfs: initialize firmware_file to empty string

In reply to Gui-Dong Han

Passing NULL to debugfs_create_str() causes a NULL pointer dereference, and creating debugfs nodes with NULL string pointers is no longer permitted.

Additionally, firmware_file is a global pointer. Previously, adding every new slave blindly overwrote it with NULL.

Fix these issues by initializing firmware_file to an allocated empty string once in the subsystem init path (sdw_debugfs_init), and freeing it in the exit path. Existing driver code handles empty strings correctly.

Fixes fe46d2a4301d ("soundwire: debugfs: add interface to read/write commands")

---
@SoundWire maintainers: Reviewed-by and Acked-by tags are welcome. Based on my testing, reading a string node created with a NULL pointer causes a crash, and writing to it returns -EINVAL. This completely breaks the interface, making me highly suspect this code has never actually been used. Additionally, sharing the global firmware_file pointer is inherently racy. I will investigate fixing or removing it entirely in a follow-up patch, as it falls outside the scope of this series.

v2:
* Replace devm_kstrdup() with kstrdup() to fix allocation/free mismatch with debugfs.
* Move initialization to sdw_debugfs_init() to correctly handle the global pointer and avoid overwriting during slave probe.

---
 drivers/soundwire/debugfs.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/soundwire/debugfs.c +7 -2
index ccc9670ef77c..2905ec19b838 100644
@@ -358,8 +358,8 @@ void sdw_slave_debugfs_init(struct sdw_slave *slave)
 	debugfs_create_file("go", 0200, d, slave, &cmd_go_fops);
 
 	debugfs_create_file("read_buffer", 0400, d, slave, &read_buffer_fops);
-	firmware_file = NULL;
-	debugfs_create_str("firmware_file", 0200, d, &firmware_file);
+	if (firmware_file)
+		debugfs_create_str("firmware_file", 0200, d, &firmware_file);
 
 	slave->debugfs = d;
 }
@@ -371,10 +371,15 @@ void sdw_slave_debugfs_exit(struct sdw_slave *slave)
 
 void sdw_debugfs_init(void)
 {
+	if (!firmware_file)
+		firmware_file = kstrdup("", GFP_KERNEL);
+
 	sdw_debugfs_root = debugfs_create_dir("soundwire", NULL);
 }
 
 void sdw_debugfs_exit(void)
 {
 	debugfs_remove_recursive(sdw_debugfs_root);
+	kfree(firmware_file);
+	firmware_file = NULL;
 }
Signature · 1 line
2.43.0