[PATCH] Bluetooth: fix the erroneous flush_work() order

From: linma <redacted>

In the cleanup routine for failed initialization of HCI device,
the flush_work(&hdev->rx_work) need to be finished before the
flush_work(&hdev->cmd_work). Otherwise, the hci_rx_work() can
possibly invoke new cmd_work and cause a bug, like double free,
in late processings.

This was assigned CVE-2021-3564.

This patch reorder the flush_work() to fix this bug.

Cc: Marcel Holtmann <marcel@holtmann.org>
Cc: Johan Hedberg <redacted>
Cc: Luiz Augusto von Dentz <luiz.dentz@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: linux-bluetooth@vger.kernel.org
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Lin Ma <redacted>
Signed-off-by: Hao Xiong <redacted>
Cc: stable <redacted>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/bluetooth/hci_core.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index fd12f1652bdf..88aa32f44e68 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c

@@ -1610,8 +1610,13 @@ static int hci_dev_do_open(struct hci_dev *hdev)
 	} else {
 		/* Init failed, cleanup */
 		flush_work(&hdev->tx_work);
-		flush_work(&hdev->cmd_work);
+		/*
+		 * Since hci_rx_work() is possible to awake new cmd_work
+		 * it should be flushed first to avoid unexpected call of
+		 * hci_cmd_work()
+		 */
 		flush_work(&hdev->rx_work);
+		flush_work(&hdev->cmd_work);
 
 		skb_queue_purge(&hdev->cmd_q);
 		skb_queue_purge(&hdev->rx_q);

-- 
2.31.1

Hi Greg,

From: linma <redacted>

this needs a real name, but I could fix that on git am as well.

So everything in net/ uses the comment coding style enforced with --strict.

Regards

Marcel

On Wed, May 26, 2021 at 05:05:50PM +0200, Marcel Holtmann wrote:

Hi Greg,

quoted

From: linma <redacted>

this needs a real name, but I could fix that on git am as well.

"Lin Ma"

quoted

In the cleanup routine for failed initialization of HCI device,
the flush_work(&hdev->rx_work) need to be finished before the
flush_work(&hdev->cmd_work). Otherwise, the hci_rx_work() can
possibly invoke new cmd_work and cause a bug, like double free,
in late processings.

This was assigned CVE-2021-3564.

This patch reorder the flush_work() to fix this bug.

Cc: Marcel Holtmann <marcel@holtmann.org>
Cc: Johan Hedberg <redacted>
Cc: Luiz Augusto von Dentz <luiz.dentz@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: linux-bluetooth@vger.kernel.org
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Lin Ma <redacted>
Signed-off-by: Hao Xiong <redacted>
Cc: stable <redacted>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/bluetooth/hci_core.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index fd12f1652bdf..88aa32f44e68 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c

@@ -1610,8 +1610,13 @@ static int hci_dev_do_open(struct hci_dev *hdev)

	} else {
		/* Init failed, cleanup */
		flush_work(&hdev->tx_work);
-		flush_work(&hdev->cmd_work);
+		/*
+		 * Since hci_rx_work() is possible to awake new cmd_work
+		 * it should be flushed first to avoid unexpected call of
+		 * hci_cmd_work()
+		 */

So everything in net/ uses the comment coding style enforced with --strict.

See v2 please.

thanks,

greg k-h