[PATCH v3] Bluetooth: eir: Fix OOB read in eir_get_service_data()

[PATCH v3] Bluetooth: eir: Fix OOB read in eir_get_service_data()

[HyeongJun An]

eir_get_service_data() walks the advertising data for a Service Data
field with a matching UUID.  On a mismatch it advances:

    eir += dlen;
    eir_len -= dlen;

eir_get_data() reports dlen as the field's data length, but the field
spans dlen + 2 bytes once its length and type bytes count, and more
when non-Service-Data fields were skipped to reach it.  The pointer
lands correctly on the next field.  eir_len does not, and the shortfall
compounds across fields until eir_get_data() reads the length and type
bytes of a "field" past the end of the buffer.

For an ISO broadcast sink that buffer is hcon->le_per_adv_data[], filled
from the periodic advertising reports of a remote broadcaster.  A PA
payload packed with mismatching Service Data fields walks off the array
into the rest of struct hci_conn.  A drifted field that matches the BAA
UUID puts those bytes in iso_pi(sk)->base, where user space reads them
back with getsockopt(BT_ISO_BASE).

Recompute eir_len from the end of the buffer each iteration.

Fixes: 8f9ae5b3ae80 ("Bluetooth: eir: Add helpers for managing service data")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-5
Signed-off-by: HyeongJun An <sammiee5311@gmail.com>
---
Changes in v3:
- Commit message only, shortened.  No code change from v2.

v2: https://lore.kernel.org/all/20260624143222.883120-1-sammiee5311@gmail.com/

v2 drew no comments.  Its bot run was 20/22 PASS.  The mgmt-tester and
mesh-tester failures are not from this patch -- eir_get_service_data()
has no callers in mesh or mgmt code, only in hci_sync.c and iso.c, and
iso-tester, which does cover the path, passed.  The previous fix to this
function, 20a2aa01f5ae ("Bluetooth: Fix NULL pointer deference on
eir_get_service_data"), also failed mgmt-tester on its bot run and was
merged.

 net/bluetooth/eir.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/bluetooth/eir.c b/net/bluetooth/eir.c
index 1de5f9df6eec..a55696820b22 100644
--- a/net/bluetooth/eir.c
+++ b/net/bluetooth/eir.c
@@ -369,6 +369,7 @@ u8 eir_create_scan_rsp(struct hci_dev *hdev, u8 instance, u8 *ptr)
 
 void *eir_get_service_data(u8 *eir, size_t eir_len, u16 uuid, size_t *len)
 {
+	const u8 *eir_end = eir + eir_len;
 	size_t dlen;
 
 	while ((eir = eir_get_data(eir, eir_len, EIR_SERVICE_DATA, &dlen))) {
@@ -381,7 +382,7 @@ void *eir_get_service_data(u8 *eir, size_t eir_len, u16 uuid, size_t *len)
 		}
 
 		eir += dlen;
-		eir_len -= dlen;
+		eir_len = eir_end - eir;
 	}
 
 	return NULL;
-- 
2.43.0

Re: [PATCH v3] Bluetooth: eir: Fix OOB read in eir_get_service_data()

[patchwork-bot+bluetooth]

Hello:

This patch was applied to bluetooth/bluetooth-next.git (master)
by Luiz Augusto von Dentz <luiz.von.dentz@intel.com>:

On Sat, 15 Aug 2026 15:24:19 +0900 you wrote:
> eir_get_service_data() walks the advertising data for a Service Data
> field with a matching UUID.  On a mismatch it advances:
> 
>     eir += dlen;
>     eir_len -= dlen;
> 
> eir_get_data() reports dlen as the field's data length, but the field
> spans dlen + 2 bytes once its length and type bytes count, and more
> when non-Service-Data fields were skipped to reach it.  The pointer
> lands correctly on the next field.  eir_len does not, and the shortfall
> compounds across fields until eir_get_data() reads the length and type
> bytes of a "field" past the end of the buffer.
> 
> [...]

Here is the summary with links:
  - [v3] Bluetooth: eir: Fix OOB read in eir_get_service_data()
    https://git.kernel.org/bluetooth/bluetooth-next/c/14a97a38ba8f

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html