[v3,1/2] Bluetooth: hci_sync: Avoid use-after-free in dbg for hci_remove_adv_monitor()

Commit Message

Doug Anderson June 30, 2023, 10:33 p.m. UTC

KASAN reports that there's a use-after-free in
hci_remove_adv_monitor(). Trawling through the disassembly, you can
see that the complaint is from the access in bt_dev_dbg() under the
HCI_ADV_MONITOR_EXT_MSFT case. The problem case happens because
msft_remove_monitor() can end up freeing the monitor
structure. Specifically:
  hci_remove_adv_monitor() ->
  msft_remove_monitor() ->
  msft_remove_monitor_sync() ->
  msft_le_cancel_monitor_advertisement_cb() ->
  hci_free_adv_monitor()

Let's fix the problem by just stashing the relevant data when it's
still valid.

Fixes: 7cf5c2978f23 ("Bluetooth: hci_sync: Refactor remove Adv Monitor")
Signed-off-by: Douglas Anderson <[e-mail]>
---

Changes in v3:
- Back to v1 where we stash the handle in a local.

Changes in v2:
- Move the print, don't stash handle in a local.

 net/bluetooth/hci_core.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

Comments

[e-mail] June 30, 2023, 10:59 p.m. UTC | #1

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=761691

---Test result---

Test Summary:
CheckPatch                    PASS      1.10 seconds
GitLint                       FAIL      0.77 seconds
SubjectPrefix                 PASS      0.16 seconds
BuildKernel                   PASS      32.78 seconds
CheckAllWarning               PASS      35.74 seconds
CheckSparse                   PASS      40.37 seconds
CheckSmatch                   PASS      112.19 seconds
BuildKernel32                 PASS      31.17 seconds
TestRunnerSetup               PASS      447.30 seconds
TestRunner_l2cap-tester       PASS      13.11 seconds
TestRunner_iso-tester         PASS      23.69 seconds
TestRunner_bnep-tester        PASS      5.58 seconds
TestRunner_mgmt-tester        PASS      132.56 seconds
TestRunner_rfcomm-tester      PASS      9.05 seconds
TestRunner_sco-tester         PASS      10.42 seconds
TestRunner_ioctl-tester       PASS      9.73 seconds
TestRunner_mesh-tester        PASS      7.02 seconds
TestRunner_smp-tester         PASS      8.21 seconds
TestRunner_userchan-tester    PASS      5.80 seconds
IncrementalBuild              PASS      35.20 seconds

Details
##############################
Test: GitLint - FAIL
Desc: Run gitlint
Output:
[v3,1/2] Bluetooth: hci_sync: Avoid use-after-free in dbg for hci_remove_adv_monitor()

WARNING: I3 - ignore-body-lines: gitlint will be switching from using Python regex 'match' (match beginning) to 'search' (match anywhere) semantics. Please review your ignore-body-lines.regex option accordingly. To remove this warning, set general.regex-style-search=True. More details: https://jorisroovers.github.io/gitlint/configuration/#regex-style-search
1: T1 Title exceeds max length (86>80): "[v3,1/2] Bluetooth: hci_sync: Avoid use-after-free in dbg for hci_remove_adv_monitor()"


---
Regards,
Linux Bluetooth

Doug Anderson June 30, 2023, 11:39 p.m. UTC | #2

Hi,

On Fri, Jun 30, 2023 at 3:59 PM <[e-mail]> wrote:
>
> 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=761691
>
> ---Test result---
>
> Test Summary:
> CheckPatch                    PASS      1.10 seconds
> GitLint                       FAIL      0.77 seconds
> SubjectPrefix                 PASS      0.16 seconds
> BuildKernel                   PASS      32.78 seconds
> CheckAllWarning               PASS      35.74 seconds
> CheckSparse                   PASS      40.37 seconds
> CheckSmatch                   PASS      112.19 seconds
> BuildKernel32                 PASS      31.17 seconds
> TestRunnerSetup               PASS      447.30 seconds
> TestRunner_l2cap-tester       PASS      13.11 seconds
> TestRunner_iso-tester         PASS      23.69 seconds
> TestRunner_bnep-tester        PASS      5.58 seconds
> TestRunner_mgmt-tester        PASS      132.56 seconds
> TestRunner_rfcomm-tester      PASS      9.05 seconds
> TestRunner_sco-tester         PASS      10.42 seconds
> TestRunner_ioctl-tester       PASS      9.73 seconds
> TestRunner_mesh-tester        PASS      7.02 seconds
> TestRunner_smp-tester         PASS      8.21 seconds
> TestRunner_userchan-tester    PASS      5.80 seconds
> IncrementalBuild              PASS      35.20 seconds
>
> Details
> ##############################
> Test: GitLint - FAIL
> Desc: Run gitlint
> Output:
> [v3,1/2] Bluetooth: hci_sync: Avoid use-after-free in dbg for hci_remove_adv_monitor()
>
> WARNING: I3 - ignore-body-lines: gitlint will be switching from using Python regex 'match' (match beginning) to 'search' (match anywhere) semantics. Please review your ignore-body-lines.regex option accordingly. To remove this warning, set general.regex-style-search=True. More details: https://jorisroovers.github.io/gitlint/configuration/#regex-style-search
> 1: T1 Title exceeds max length (86>80): "[v3,1/2] Bluetooth: hci_sync: Avoid use-after-free in dbg for hci_remove_adv_monitor()"

I'll plan to ignore this one. It looks like it only exceeds 80
characters because of the version prefix "[v3, 1/2] ". If someone
really wants me to re-post with a shorter subject, please yell and I
will do that.

Thanks!

-Doug

[e-mail] July 5, 2023, 11 p.m. UTC | #3

Hello:

This series was applied to bluetooth/bluetooth-next.git (master)
by Luiz Augusto von Dentz <[e-mail]>:

On Fri, 30 Jun 2023 15:33:14 -0700 you wrote:
> KASAN reports that there's a use-after-free in
> hci_remove_adv_monitor(). Trawling through the disassembly, you can
> see that the complaint is from the access in bt_dev_dbg() under the
> HCI_ADV_MONITOR_EXT_MSFT case. The problem case happens because
> msft_remove_monitor() can end up freeing the monitor
> structure. Specifically:
>   hci_remove_adv_monitor() ->
>   msft_remove_monitor() ->
>   msft_remove_monitor_sync() ->
>   msft_le_cancel_monitor_advertisement_cb() ->
>   hci_free_adv_monitor()
> 
> [...]

Here is the summary with links:
  - [v3,1/2] Bluetooth: hci_sync: Avoid use-after-free in dbg for hci_remove_adv_monitor()
    https://git.kernel.org/bluetooth/bluetooth-next/c/2648c5eb33ca
  - [v3,2/2] Bluetooth: hci_sync: Don't double print name in add/remove adv_monitor
    https://git.kernel.org/bluetooth/bluetooth-next/c/22d2055a576d

You are awesome, thank you!

Patch

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 48917c68358d..dbb2043a9112 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -1972,6 +1972,7 @@  static int hci_remove_adv_monitor(struct hci_dev *hdev,
 				  struct adv_monitor *monitor)
 {
 	int status = 0;
+	int handle;
 
 	switch (hci_get_adv_monitor_offload_ext(hdev)) {
 	case HCI_ADV_MONITOR_EXT_NONE: /* also goes here when powered off */
@@ -1980,9 +1981,10 @@  static int hci_remove_adv_monitor(struct hci_dev *hdev,
 		goto free_monitor;
 
 	case HCI_ADV_MONITOR_EXT_MSFT:
+		handle = monitor->handle;
 		status = msft_remove_monitor(hdev, monitor);
 		bt_dev_dbg(hdev, "%s remove monitor %d msft status %d",
-			   hdev->name, monitor->handle, status);
+			   hdev->name, handle, status);
 		break;
 	}