MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date:   Thu, 07 Mar 2019 12:19:44 +0530
From:   Balakrishna Godavarthi <bgodavar@codeaurora.org>
To:     Stephen Boyd <swboyd@chromium.org>
Cc:     johan.hedberg@gmail.com, marcel@holtmann.org, mka@chromium.org,
        linux-kernel@vger.kernel.org, linux-bluetooth@vger.kernel.org,
        hemantg@codeaurora.org, linux-arm-msm@vger.kernel.org,
        rjliao@codeaurora.org
Subject: Re: [PATCH v1] Bluetooth: hci_qca: Give enough time to ROME
 controller to bootup.
In-Reply-To: <155191158734.20095.15990837191526416248@swboyd.mtv.corp.google.com>
References: <20190306162113.8585-1-bgodavar@codeaurora.org>
 <155191158734.20095.15990837191526416248@swboyd.mtv.corp.google.com>
Message-ID: <a22894cea091ee9945af9c22d6e42051@codeaurora.org>
User-Agent: Roundcube Webmail/1.2.5
Sender: linux-bluetooth-owner@vger.kernel.org
Precedence: bulk

Hi Stepen,

On 2019-03-07 04:03, Stephen Boyd wrote:
> Quoting Balakrishna Godavarthi (2019-03-06 08:21:13)
>> This patch enables enough time to ROME controller to bootup
>> after we bring the enable ping out of reset.
>> 
>> Signed-off-by: Balakrishna Godavarthi <bgodavar@codeaurora.org>
>> ---
> 
> Any Fixes tag? And maybe some more explanation or background on where
> 150 ms sleep comes from would be useful. Was it determined
> experimentally or did it come from a datasheet somewhere? Does the time
> differ between boards?
> 
[Bala]: this was observed in our stress testing and even the CHIP 
firmware team
         confirmed that BT chip required at least 150 ms to boot up.

@Rocky to confirm my statement.

>>  drivers/bluetooth/hci_qca.c | 2 ++
>>  1 file changed, 2 insertions(+)
>> 
>> diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
>> index 237aea34b69f..1953b13511e7 100644
>> --- a/drivers/bluetooth/hci_qca.c
>> +++ b/drivers/bluetooth/hci_qca.c
>> @@ -508,6 +508,8 @@ static int qca_open(struct hci_uart *hu)
>>                 qcadev = serdev_device_get_drvdata(hu->serdev);
>>                 if (qcadev->btsoc_type != QCA_WCN3990) {
>>                         gpiod_set_value_cansleep(qcadev->bt_en, 1);
>> +                       /* Controller needs time to bootup. */
>> +                       msleep(150);
>>                 } else {
>>                         hu->init_speed = qcadev->init_speed;
>>                         hu->oper_speed = qcadev->oper_speed;

-- 
Regards
Balakrishna.