On Tue, May 02, 2023 at 04:44:48PM -0700, Luiz Augusto von Dentz wrote:

Hi Johan,

On Mon, Apr 24, 2023 at 6:35 AM Johan Hovold [off-list ref] wrote:

quoted

Devices that lack persistent storage for the device address can indicate
this by setting the HCI_QUIRK_INVALID_BDADDR which causes the controller
to be marked as unconfigured until user space has set a valid address.

Once configured, the device address must be set on every setup for
controllers with HCI_QUIRK_NON_PERSISTENT_SETUP to avoid marking the
controller as unconfigured and requiring the address to be set again.

Fixes: 740011cfe948 ("Bluetooth: Add new quirk for non-persistent setup settings")
Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
---
 net/bluetooth/hci_sync.c | 28 +++++++++++-----------------
 1 file changed, 11 insertions(+), 17 deletions(-)

diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index a8785126df75..f45598b5a532 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c

@@ -4573,23 +4573,17 @@ static int hci_dev_setup_sync(struct hci_dev *hdev)
        invalid_bdaddr = test_bit(HCI_QUIRK_INVALID_BDADDR, &hdev->quirks);

        if (!ret) {
-               if (test_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks)) {
-                       if (!bacmp(&hdev->public_addr, BDADDR_ANY))
-                               hci_dev_get_bd_addr_from_property(hdev);
-
-                       if (bacmp(&hdev->public_addr, BDADDR_ANY) &&
-                           hdev->set_bdaddr) {
-                               ret = hdev->set_bdaddr(hdev,
-                                                      &hdev->public_addr);
-
-                               /* If setting of the BD_ADDR from the device
-                                * property succeeds, then treat the address
-                                * as valid even if the invalid BD_ADDR
-                                * quirk indicates otherwise.
-                                */
-                               if (!ret)
-                                       invalid_bdaddr = false;
-                       }
+               if (test_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks) &&
+                   !bacmp(&hdev->public_addr, BDADDR_ANY))
+                       hci_dev_get_bd_addr_from_property(hdev);
+
+               if ((invalid_bdaddr ||
+                    test_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks)) &&
+                   bacmp(&hdev->public_addr, BDADDR_ANY) &&
+                   hdev->set_bdaddr) {
+                       ret = hdev->set_bdaddr(hdev, &hdev->public_addr);
+                       if (!ret)
+                               invalid_bdaddr = false;

I'd keep the original comments since it appears you haven't changed
its logic with respect to invalid_bdaddr.

The comment no longer applies as set_bdaddr can now be called also when
there is no "device property" holding the address (e.g. when one has
been set using btmgmt and HCI_QUIRK_NON_PERSISTENT_SETUP is set).

And obviously we shouldn't mark the controller as unconfigured after
we've successfully set an address.

Johan