[PATCH bluetooth] Bluetooth: hci_sync: take hci_dev_lock in hci_clear_adv_instance_sync

[PATCH bluetooth] Bluetooth: hci_sync: take hci_dev_lock in hci_clear_adv_instance_sync

[Xiang Mei]

hci_clear_adv_instance_sync() runs on the cmd_sync worker without
hdev->lock, yet walks hdev->adv_instances and calls
hci_remove_adv_instance()/hci_find_adv_instance()/hci_get_next_instance(),
all of which require hdev->lock. The list is otherwise serialized by
hdev->lock (add/remove_advertising mutate it, read_adv_features()
traverses it), so a concurrent holder can list_del()+kfree() or list_add()
an adv_info node mid-traversal, causing a use-after-free and list
corruption. Reachable by racing MGMT_OP_SET_LE(off) or advertising timeout
expiry against MGMT add/remove/read advertising:

  BUG: KASAN: slab-use-after-free in read_adv_features+0x44b/0x500
  Read of size 8 at addr ffff888013878000 by task exploit/150
  ...
  Call Trace:
   read_adv_features (net/bluetooth/mgmt.c:8546)
   hci_sock_sendmsg (net/bluetooth/hci_sock.c:1719)
   sock_write_iter (net/socket.c:1241)
   vfs_write (fs/read_write.c:687)
   ksys_write (fs/read_write.c:739)
   do_syscall_64 (arch/x86/entry/syscall_64.c:94)
  ...
  Freed by task 52:
  ...
  Oops: general protection fault, probably for non-canonical address
  RIP: 0010:read_adv_features (net/bluetooth/mgmt.c:8548)

Hold hdev->lock around the traversal. Both callers run on the worker
without hdev->lock, so this cannot deadlock.

Fixes: c249ea9b4309 ("Bluetooth: Move Adv Instance timer to hci_sync")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Xiang Mei <xmei5@asu.edu>
---
 net/bluetooth/hci_sync.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 3be8c3581c6c..6b566c67ee98 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -487,6 +487,8 @@ int hci_clear_adv_instance_sync(struct hci_dev *hdev, struct sock *sk,
 	int err;
 	u8 rem_inst;
 
+	hci_dev_lock(hdev);
+
 	/* Cancel any timeout concerning the removed instance(s). */
 	if (!instance || hdev->cur_adv_instance == instance)
 		cancel_adv_timeout(hdev);
@@ -525,6 +527,8 @@ int hci_clear_adv_instance_sync(struct hci_dev *hdev, struct sock *sk,
 		}
 	}
 
+	hci_dev_unlock(hdev);
+
 	if (!hdev_is_powered(hdev) || hci_dev_test_flag(hdev, HCI_ADVERTISING))
 		return 0;
 
-- 
2.43.0

RE: Bluetooth: hci_sync: take hci_dev_lock in hci_clear_adv_instance_sync

[bluez.test.bot]

[-- Attachment #1: Type: text/plain, Size: 3535 bytes --]

This is automated email and please do not reply to this email!

Dear submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=1121363

---Test result---

Test Summary:
CheckPatch                    FAIL      0.72 seconds
VerifyFixes                   PASS      0.13 seconds
VerifySignedoff               PASS      0.13 seconds
GitLint                       PASS      0.32 seconds
SubjectPrefix                 PASS      0.12 seconds
BuildKernel                   PASS      27.14 seconds
CheckAllWarning               PASS      29.13 seconds
CheckSparse                   PASS      27.88 seconds
BuildKernel32                 PASS      25.90 seconds
CheckKernelLLVM               SKIP      0.00 seconds
TestRunnerSetup               PASS      492.72 seconds
TestRunner_l2cap-tester       FAIL      58.77 seconds
TestRunner_iso-tester         PASS      88.28 seconds
TestRunner_bnep-tester        PASS      18.79 seconds
TestRunner_mgmt-tester        FAIL      210.08 seconds
TestRunner_rfcomm-tester      PASS      25.03 seconds
TestRunner_sco-tester         PASS      34.71 seconds
TestRunner_ioctl-tester       PASS      25.82 seconds
TestRunner_mesh-tester        FAIL      25.88 seconds
TestRunner_smp-tester         PASS      23.31 seconds
TestRunner_userchan-tester    PASS      19.42 seconds
TestRunner_6lowpan-tester     PASS      22.11 seconds
IncrementalBuild              PASS      25.85 seconds

Details
##############################
Test: CheckPatch - FAIL
Desc: Run checkpatch.pl script
Output:
Bluetooth: hci_sync: take hci_dev_lock in hci_clear_adv_instance_sync
WARNING: Reported-by: should be immediately followed by Closes: with a URL to the report
#129: 
Reported-by: Weiming Shi <bestswngs@gmail.com>
Assisted-by: Claude:claude-opus-4-8

total: 0 errors, 1 warnings, 0 checks, 16 lines checked

NOTE: For some of the reported defects, checkpatch may be able to
      mechanically convert to the typical style using --fix or --fix-inplace.

/github/workspace/src/patch/14663493.patch has style problems, please review.

NOTE: Ignored message types: UNKNOWN_COMMIT_ID

NOTE: If any of the errors are false positives, please report
      them to the maintainer, see CHECKPATCH in MAINTAINERS.


##############################
Test: CheckKernelLLVM - SKIP
Desc: Build kernel with LLVM + context analysis
Output:
Clang not found
##############################
Test: TestRunner_l2cap-tester - FAIL
Desc: Run l2cap-tester with test-runner
Output:
Total: 96, Passed: 94 (97.9%), Failed: 2, Not Run: 0

Failed Test Cases
L2CAP BR/EDR Server - Set PHY 1M                     Failed       0.264 seconds
L2CAP BR/EDR Server - Set PHY 2M                     Failed       0.265 seconds
##############################
Test: TestRunner_mgmt-tester - FAIL
Desc: Run mgmt-tester with test-runner
Output:
Total: 494, Passed: 489 (99.0%), Failed: 1, Not Run: 4

Failed Test Cases
Read Exp Feature - Success                           Failed       0.238 seconds
##############################
Test: TestRunner_mesh-tester - FAIL
Desc: Run mesh-tester with test-runner
Output:
Total: 10, Passed: 8 (80.0%), Failed: 2, Not Run: 0

Failed Test Cases
Mesh - Send cancel - 1                               Timed out    2.669 seconds
Mesh - Send cancel - 2                               Timed out    1.987 seconds


https://github.com/bluez/bluetooth-next/pull/395

---
Regards,
Linux Bluetooth