[PATCH v2] klist: avoid accesses after waking klist_remove()

[Karl Mehltretter]

klist_remove() waits until a node is unreferenced so that its caller can
free the containing object.  klist_release() currently publishes
waiter->woken and wakes the waiter before its final accesses to the waiter
and node.  klist_remove() can then return, allowing its stack waiter and
the containing object to be freed or reused while klist_release() is still
running.  In particular, bus_remove_driver() can free drv->p while
__device_attach() walks the same bus klist with bus_for_each_drv().

On an arm64 Cortex-A72 system, an unpatched 7.2.0-rc3 kernel with
CONFIG_PREEMPT_RT=y and CONFIG_KASAN=y reproduced the bug through the
in-tree I2C/at24 path.  KASAN reported a use-after-free in
klist_dec_and_del() reached from klist_next()/bus_for_each_drv() while
at24 was being unregistered.

Clear n_klist and take a task reference before publishing woken.  Use
release/acquire accesses for that publication and wake the referenced task.
The task reference keeps the waiter task alive if it returns and exits
before wake_up_process().

Fixes: 8b0c250be489 ("[PATCH] add klist_node_attached() to determine if a node is on a list or not.")
Fixes: 210272a28465 ("driver core: Remove completion from struct klist_node")
Cc: stable@xxxxxxxxxxxxxxx
Assisted-by: LLM
Signed-off-by: Karl Mehltretter <kmehltretter@xxxxxxxxx>
---
v2:
- Explain the use-after-free and document an in-tree driver path; drop the
  syzbot tags (Greg).
- Keep the mb(); removing it was NAKed before:
  https://lore.kernel.org/r/20220614144443.6566-1-wuchi.zero@xxxxxxxxx/
- Add an arm64 Cortex-A72 PREEMPT_RT + KASAN reproduction through the
  in-tree I2C/at24 driver path; the first run hit KASAN after 32 module
  unregisters and 397 device operations.

Link to v1:
  https://lore.kernel.org/r/20260825040408.59225-1-kmehltretter@xxxxxxxxx/

 lib/klist.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/lib/klist.c b/lib/klist.c
index 332a4fbf18ff0..f133740b1c2cb 100644
--- a/lib/klist.c
+++ b/lib/klist.c
@@ -36,6 +36,7 @@
 #include <linux/klist.h>
 #include <linux/export.h>
 #include <linux/sched.h>
+#include <linux/sched/task.h>
 
 /*
  * Use the lowest bit of n_klist to mark deleted nodes and exclude
@@ -187,18 +188,24 @@ static void klist_release(struct kref *kref)
 
 	WARN_ON(!knode_dead(n));
 	list_del(&n->n_node);
+	knode_set_klist(n, NULL);
 	spin_lock(&klist_remove_lock);
 	list_for_each_entry_safe(waiter, tmp, &klist_remove_waiters, list) {
+		struct task_struct *p;
+
 		if (waiter->node != n)
 			continue;
 
+		p = waiter->process;
+		get_task_struct(p);
 		list_del(&waiter->list);
-		waiter->woken = 1;
+		/* Publish only after the final waiter and n accesses */
+		smp_store_release(&waiter->woken, 1);
 		mb();
-		wake_up_process(waiter->process);
+		wake_up_process(p);
+		put_task_struct(p);
 	}
 	spin_unlock(&klist_remove_lock);
-	knode_set_klist(n, NULL);
 }
 
 static int klist_dec_and_del(struct klist_node *n)
@@ -250,7 +257,8 @@ void klist_remove(struct klist_node *n)
 
 	for (;;) {
 		set_current_state(TASK_UNINTERRUPTIBLE);
-		if (waiter.woken)
+		/* Pairs with the release store in klist_release() */
+		if (smp_load_acquire(&waiter.woken))
 			break;
 		schedule();
 	}
-- 
2.53.0