Received-SPF: pass (google.com: domain of linux-bluetooth+bounces-3630-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <20240412084128.24452-1-hao.qin@mediatek.com>
In-Reply-To: <20240412084128.24452-1-hao.qin@mediatek.com>
From: Luiz Augusto von Dentz <luiz.dentz@gmail.com>
Date: Tue, 16 Apr 2024 10:53:43 -0400
Message-ID: <CABBYNZLGocaksAKCTso3zL-bAuSKaBurTtOFZx_rw=HiKMmqtA@mail.gmail.com>
Subject: Re: [PATCH] Bluetooth: btusb: mediatek: Add usb reset for firmware
 setup failure
To: Hao Qin <hao.qin@mediatek.com>
Cc: Marcel Holtmann <marcel@holtmann.org>, Johan Hedberg <johan.hedberg@gmail.com>, 
	Sean Wang <sean.wang@mediatek.com>, Deren Wu <deren.Wu@mediatek.com>, 
	Aaron Hou <aaron.hou@mediatek.com>, Chris Lu <chris.lu@mediatek.com>, 
	Steve Lee <steve.lee@mediatek.com>, linux-bluetooth <linux-bluetooth@vger.kernel.org>, 
	linux-kernel <linux-kernel@vger.kernel.org>, 
	linux-mediatek <linux-mediatek@lists.infradead.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi,

On Fri, Apr 12, 2024 at 4:43=E2=80=AFAM Hao Qin <hao.qin@mediatek.com> wrot=
e:
>
> Add usb reset to recover BT device from a firmware setup failure,
> it will restore the device to its original state and attempt setup
> firmware again.
>
> Signed-off-by: Hao Qin <hao.qin@mediatek.com>
> ---
>  drivers/bluetooth/btusb.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
> index 4c0cc13adb47..c5aab361795e 100644
> --- a/drivers/bluetooth/btusb.c
> +++ b/drivers/bluetooth/btusb.c
> @@ -3160,6 +3160,9 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
>                                                 btusb_mtk_hci_wmt_sync);
>                 if (err < 0) {
>                         bt_dev_err(hdev, "Failed to set up firmware (%d)"=
, err);
> +                       btusb_stop_traffic(data);
> +                       usb_kill_anchored_urbs(&data->tx_anchor);
> +                       usb_queue_reset_device(data->intf);

What if this was already caused by a reset? Wouldn't that create a
loop resetting the device over and over again?

>                         return err;
>                 }
>
> --
> 2.18.0
>


--=20
Luiz Augusto von Dentz