Hi Amit,
> In case of BT turn off from UI it is safe to add HCI reset command
> which will maintain the controller state.
this needs to be a bit more details since hdev->shutdown handling is really only for the older Intel controllers.
Also I am failing to see why HCI Reset helps or what it does here. It is going to be called again on hdev->open the next time the controller is activated.
>
> Signed-off-by: Amit K Bag <amit.k.bag@xxxxxxxxx>
> ---
> drivers/bluetooth/btusb.c | 12 ++++++++++++
> 1 file changed, 12 insertions(+)
>
> diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
> index 572fd75fbcf6..599ec9b12d3f 100644
> --- a/drivers/bluetooth/btusb.c
> +++ b/drivers/bluetooth/btusb.c
> @@ -2369,6 +2369,18 @@ static int btusb_shutdown_intel(struct hci_dev *hdev)
> struct sk_buff *skb;
> long ret;
>
> + /* In case of BT off from UI it is safe to do HCI reset.
> + * This will maintain the controller state.
> + */
I do not care what the UI does here. This is power down operation and so it needs to be explained properly. And with a lot more details since I am failing to understand what state is maintained on HCI Reset. As per specification it will reset all Bluetooth states back to the defaults.
> + skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_INIT_TIMEOUT);
> + if (IS_ERR(skb)) {
> + ret = PTR_ERR(skb);
> + BT_ERR("%s: BT controller HCI reset failed (%ld)",
> + hdev->name, ret);
> + return ret;
> + }
> + kfree_skb(skb);
> +
There is whitespace damage here. And if this hasn’t been done yet, please first start using bt_dev_err instead of BT_ERR for the Intel parts.
> /* Some platforms have an issue with BT LED when the interface is
> * down or BT radio is turned off, which takes 5 seconds to BT LED
> * goes off. This command turns off the BT LED immediately.
Regards
Marcel
--
To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
