Skip to contentSkip to conversation

kernfs: Relax constraint in draining guard

PATCH In review Michal Koutný 3 participants 5 messages May 5, 2025 – May 14, 2025 kernfs

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 kernfs: Relax constraint in draining guard

The active reference lifecycle provides the break/unbreak mechanism but the active reference is not truly active after unbreak -- callers don't use it afterwards but it's important for proper pairing of kn->active counting. Assuming this mechanism is in place, the WARN check in kernfs_should_drain_open_files() is too sensitive -- it may transiently catch those (rightful) callers between kernfs_unbreak_active_protection() and kernfs_put_active() as found out by Chen Ridong:

	kernfs_remove_by_name_ns	kernfs_get_active // active=1
	__kernfs_remove					  // active=0x80000002
	kernfs_drain			...
	wait_event
	//waiting (active == 0x80000001)
					kernfs_break_active_protection
					// active = 0x80000001
	// continue
					kernfs_unbreak_active_protection
					// active = 0x80000002
	...
	kernfs_should_drain_open_files
	// warning occurs
					kernfs_put_active

To avoid the false positives (mind panic_on_warn) remove the check altogether. (This is meant as quick fix, I think active reference break/unbreak may be simplified with larger rework.)

Fixes bdb2fd7fc56e1 ("kernfs: Skip kernfs_drain_open_files() more aggressively")
---
 fs/kernfs/dir.c  | 5 +++--
 fs/kernfs/file.c | 3 ++-
 2 files changed, 5 insertions(+), 3 deletions(-)
fs/kernfs/dir.c +3 -2
index fc70d72c3fe80..43487fa83eaea 100644
@@ -1580,8 +1580,9 @@ void kernfs_break_active_protection(struct kernfs_node *kn)
  * invoked before finishing the kernfs operation.  Note that while this
  * function restores the active reference, it doesn't and can't actually
  * restore the active protection - @kn may already or be in the process of
- * being removed.  Once kernfs_break_active_protection() is invoked, that
- * protection is irreversibly gone for the kernfs operation instance.
+ * being drained and removed.  Once kernfs_break_active_protection() is
+ * invoked, that protection is irreversibly gone for the kernfs operation
+ * instance.
  *
  * While this function may be called at any point after
  * kernfs_break_active_protection() is invoked, its most useful location
fs/kernfs/file.c +2 -1
index 66fe8fe41f060..a6c692cac6165 100644
@@ -778,8 +778,9 @@ bool kernfs_should_drain_open_files(struct kernfs_node *kn)
 	/*
 	 * @kn being deactivated guarantees that @kn->attr.open can't change
 	 * beneath us making the lockless test below safe.
+	 * Callers post kernfs_unbreak_active_protection may be counted in
+	 * kn->active by now, do not WARN_ON because of them.
 	 */
-	WARN_ON_ONCE(atomic_read(&kn->active) != KN_DEACTIVATED_BIAS);
 
 	rcu_read_lock();
 	on = rcu_dereference(kn->attr.open);
Signature · 1 line
2.49.0
maintainer Acked-by on v1 1/1
#1 lore raw
Quoted 28 lines · Michal Koutný

The active reference lifecycle provides the break/unbreak mechanism but the active reference is not truly active after unbreak -- callers don't use it afterwards but it's important for proper pairing of kn->active counting. Assuming this mechanism is in place, the WARN check in kernfs_should_drain_open_files() is too sensitive -- it may transiently catch those (rightful) callers between kernfs_unbreak_active_protection() and kernfs_put_active() as found out by Chen Ridong:

	kernfs_remove_by_name_ns	kernfs_get_active // active=1
	__kernfs_remove					  // active=0x80000002
	kernfs_drain			...
	wait_event
	//waiting (active == 0x80000001)
					kernfs_break_active_protection
					// active = 0x80000001
	// continue
					kernfs_unbreak_active_protection
					// active = 0x80000002
	...
	kernfs_should_drain_open_files
	// warning occurs
					kernfs_put_active

To avoid the false positives (mind panic_on_warn) remove the check altogether. (This is meant as quick fix, I think active reference break/unbreak may be simplified with larger rework.)

Quoted 5 lines
Fixes bdb2fd7fc56e1 ("kernfs: Skip kernfs_drain_open_files() more aggressively")

Thanks.

Signature · 1 line
tejun