[PATCH v2 0/2] ima: don't measure/appraise files on configfs

[PATCH v2 0/2] ima: don't measure/appraise files on configfs

[Frederick Lawler]

IMA measurement of a configfs file causes process_measurement() to hold
iint->mutex while performing a kernel_read() to hash it, which re-enters
configfs's own file locking (buffer->mutex, frag_sem).

Separately, opening any file with O_TRUNC now causes ima_file_truncate()
to take iint->mutex to reset the cached action flags, while sb_writers is
already held for that mount.

When a configfs-backed nvmet namespace is involved, these two independent
lock chains combine into a cycle:

iint->mutex -> configfs locks -> subsys->lock -> sb_writers -> iint->mutex

Add configfs to the builtin don't measure/appraise rules, similarly to
other pseudo file systems, so IMA never takes iint->mutex for configfs
file in the first place. 

Link: https://lore.kernel.org/all/6a77c7cd.b50370da.49fe0.0031.GAE@google.com/#t

Signed-off-by: Frederick Lawler <fred@cloudflare.com>
---
Changes in v2:
- Add CONFIGFS_MAGIC to IMA documentation in patch 2 (per shashiko)
- Update cover letter + commit message as suggested
- Link to v1: https://lore.kernel.org/r/20260818-configfs-v1-0-a2329043cf86@cloudflare.com

---
Frederick Lawler (2):
      configfs: move CONFIGFS_MAGIC definition to magic.h
      ima: don't measure/appraise files on configfs

 Documentation/ABI/testing/ima_policy | 3 +++
 fs/configfs/mount.c                  | 4 +---
 include/uapi/linux/magic.h           | 1 +
 security/integrity/ima/ima_policy.c  | 7 ++++++-
 4 files changed, 11 insertions(+), 4 deletions(-)
---
base-commit: f1e10b10874051e4d99911dae0dd7b75a9f8ae66
change-id: 20260818-configfs-a248c410b083

Best regards,
-- 
Frederick Lawler <fred@cloudflare.com>

[PATCH v2 1/2] configfs: move CONFIGFS_MAGIC definition to magic.h

[Frederick Lawler]

IMA shouldn't measure or appraise configfs, but currently does because
it's missing from the default exclusion policies. Move CONFIGFS_MAGIC to
magic.h to expose the file system's magic to IMA, as well as other userland
applications.

Suggested-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Frederick Lawler <fred@cloudflare.com>
---
 fs/configfs/mount.c        | 4 +---
 include/uapi/linux/magic.h | 1 +
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c
index 4929f343118946eaa55a539db4192e9c6621a8dc..d8cac1cbf3bd573a0549f2f044714c9d669205a4 100644
--- a/fs/configfs/mount.c
+++ b/fs/configfs/mount.c
@@ -9,6 +9,7 @@
  */
 
 #include <linux/fs.h>
+#include <linux/magic.h>
 #include <linux/module.h>
 #include <linux/mount.h>
 #include <linux/fs_context.h>
@@ -19,9 +20,6 @@
 #include <linux/configfs.h>
 #include "configfs_internal.h"
 
-/* Random magic number */
-#define CONFIGFS_MAGIC 0x62656570
-
 static struct vfsmount *configfs_mount = NULL;
 struct kmem_cache *configfs_dir_cachep;
 static int configfs_mnt_count = 0;
diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h
index 4f2da935a76ccae6e1c3310b2fae5379186548b8..d5b9958ddfaae90b270b66fe7853c5a3d49ce739 100644
--- a/include/uapi/linux/magic.h
+++ b/include/uapi/linux/magic.h
@@ -8,6 +8,7 @@
 #define AUTOFS_SUPER_MAGIC	0x0187
 #define CEPH_SUPER_MAGIC	0x00c36400
 #define CODA_SUPER_MAGIC	0x73757245
+#define CONFIGFS_MAGIC		0x62656570	/* some random number */
 #define CRAMFS_MAGIC		0x28cd3d45	/* some random number */
 #define CRAMFS_MAGIC_WEND	0x453dcd28	/* magic number with the wrong endianess */
 #define DEBUGFS_MAGIC          0x64626720

-- 
2.43.0

[PATCH v2 2/2] ima: don't measure/appraise files on configfs

[Frederick Lawler]

IMA measurement of a configfs file causes process_measurement() to hold
iint->mutex while performing a kernel_read() to hash it, which re-enters
configfs's own file locking (buffer->mutex, frag_sem).

Separately, opening any file with O_TRUNC now causes ima_file_truncate()
to take iint->mutex to reset the cached action flags, while sb_writers is
already held for that mount.

When a configfs-backed nvmet namespace is involved, these two independent
lock chains combine into a cycle:

iint->mutex -> configfs locks -> subsys->lock -> sb_writers -> iint->mutex

Add configfs to the builtin don't measure/appraise rules, similarly to
other pseudo file systems, so IMA never takes iint->mutex for configfs
file in the first place.

Reported-by: syzbot+448c2e24b1ceff13ed2a@syzkaller.appspotmail.com
Link: https://lore.kernel.org/all/6a77c7cd.b50370da.49fe0.0031.GAE@google.com/
Suggested-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Frederick Lawler <fred@cloudflare.com>
---
 Documentation/ABI/testing/ima_policy | 3 +++
 security/integrity/ima/ima_policy.c  | 7 ++++++-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/Documentation/ABI/testing/ima_policy b/Documentation/ABI/testing/ima_policy
index 19258471b7b26b2c42c1ab2d11b4fd630b2e6f81..b8a763e4c9fb1a5ccca97518e80e0452554c9e96 100644
--- a/Documentation/ABI/testing/ima_policy
+++ b/Documentation/ABI/testing/ima_policy
@@ -108,6 +108,9 @@ Description:
 			# NSFS_MAGIC
 			dont_measure fsmagic=0x6e736673
 			dont_appraise fsmagic=0x6e736673
+			# CONFIGFS_MAGIC
+			dont_measure fsmagic=0x62656570
+			dont_appraise fsmagic=0x62656570
 
 			measure func=BPRM_CHECK
 			measure func=FILE_MMAP mask=MAY_EXEC
diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index f79d07bb63c6fc4ba6fe594140de8d59f57e4f0b..68d9a5e6c232ea0678e9f51f105cebecccccb43e 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -165,7 +165,10 @@ static struct ima_rule_entry dont_measure_rules[] __ro_after_init = {
 	{.action = DONT_MEASURE, .fsmagic = CGROUP2_SUPER_MAGIC,
 	 .flags = IMA_FSMAGIC},
 	{.action = DONT_MEASURE, .fsmagic = NSFS_MAGIC, .flags = IMA_FSMAGIC},
-	{.action = DONT_MEASURE, .fsmagic = EFIVARFS_MAGIC, .flags = IMA_FSMAGIC}
+	{.action = DONT_MEASURE, .fsmagic = EFIVARFS_MAGIC,
+	 .flags = IMA_FSMAGIC},
+	{.action = DONT_MEASURE, .fsmagic = CONFIGFS_MAGIC,
+	 .flags = IMA_FSMAGIC}
 };
 
 static struct ima_rule_entry original_measurement_rules[] __ro_after_init = {
@@ -211,6 +214,8 @@ static struct ima_rule_entry default_appraise_rules[] __ro_after_init = {
 	{.action = DONT_APPRAISE, .fsmagic = EFIVARFS_MAGIC, .flags = IMA_FSMAGIC},
 	{.action = DONT_APPRAISE, .fsmagic = CGROUP_SUPER_MAGIC, .flags = IMA_FSMAGIC},
 	{.action = DONT_APPRAISE, .fsmagic = CGROUP2_SUPER_MAGIC, .flags = IMA_FSMAGIC},
+	{.action = DONT_APPRAISE, .fsmagic = CONFIGFS_MAGIC,
+	 .flags = IMA_FSMAGIC},
 #ifdef CONFIG_IMA_WRITE_POLICY
 	{.action = APPRAISE, .func = POLICY_CHECK,
 	.flags = IMA_FUNC | IMA_DIGSIG_REQUIRED},

-- 
2.43.0