Received-SPF: pass (google.com: domain of linux-bluetooth+bounces-4066-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Message-ID: <28f245c0-e81c-403b-ae7c-94fbd727998a@quicinc.com>
Date: Thu, 25 Apr 2024 18:28:14 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v1] Bluetooth: qca: Correct property enable-gpios handling
 logic for WCN6750 and WCN6855
To: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        Bartosz Golaszewski
	<bartosz.golaszewski@linaro.org>,
        Wren Turkal <wt@penguintechs.org>
CC: Krzysztof Kozlowski <krzk@kernel.org>, <luiz.dentz@gmail.com>,
        <luiz.von.dentz@intel.com>, <marcel@holtmann.org>,
        <linux-bluetooth@vger.kernel.org>, <kernel@quicinc.com>
References: <1713947712-4307-1-git-send-email-quic_zijuhu@quicinc.com>
 <af004bcd-0649-4a82-8f09-36d6addd0e1e@linaro.org>
 <0cba3838-917b-48af-ba8b-2ef60714943a@quicinc.com>
 <49cf0a3d-5f52-41a1-9858-692f48d06b88@kernel.org>
 <c1632816-f754-42c6-8448-2daff8221b58@penguintechs.org>
 <368e70e0-edc2-44fd-a9a2-72efa0d9f864@linaro.org>
 <f372c661-d82a-4f6f-b2de-8dd241d6b107@penguintechs.org>
 <CACMJSesk+DYFDwM7hUoPs9Xy+i1q8w+5CFu4_RdreUXudk18zw@mail.gmail.com>
 <d4c65299-d3af-4741-adc8-aafc1ab34369@linaro.org>
Content-Language: en-US
From: quic_zijuhu <quic_zijuhu@quicinc.com>
In-Reply-To: <d4c65299-d3af-4741-adc8-aafc1ab34369@linaro.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit

On 4/25/2024 5:14 PM, Krzysztof Kozlowski wrote:
> On 25/04/2024 10:30, Bartosz Golaszewski wrote:
>> On Thu, 25 Apr 2024 at 08:30, Wren Turkal <wt@penguintechs.org> wrote:
>>>
>>> On 4/24/24 11:05 PM, Krzysztof Kozlowski wrote:
>>>> On 25/04/2024 00:01, Wren Turkal wrote:
>>>>>>>
>>>>>>> 3) only care about the case property enable-gpios is not configured,
>>>>>>> the original BT driver design logic indeed matches with binging spec's
>>>>>>> requirements before bartosz's wrong change
>>>>>>
>>>>>> What? There is no such case according to bindings. I told you already
>>>>>> two times: Either change bindings or this is not valid.
>>>>>
>>>>> @krzysztof, I'm curious. There is no entry in the binding specifically
>>>>> for qca6390. Should there be?
>>>>
>>>> qca6390 is documented in the bindings, but you are right that it lacks
>>>> if:then: entry narrowing/choosing specific supplies and pins. It should
>>>> have one, indeed.
>>>
>>> Would creating an entry for the qca6390 hardware fix the issue you are
>>> worried about?
>>>
>>> Again, sorry for all the, what I assume are, basic questions. I am so
>>> far out of my depth here. I am just trying to figure out how to move
>>> forward. I really do appreciate your guidance here.
>>>
>>
>> Wren, Krzysztof: I cannot stop you from doing this but I'm afraid this
>> will complicate the power sequencing work unnecessarily. The QCA6390
>> PMU bindings I'm proposing[1] are consumers of the BT enable GPIOs. In
>> my series I also create an entry for QCA6390 Bluetooth[2] but without
>> enable-gpios and with the inputs from the PMU, not host. Please
>> consider that if you decide to go with this.
> 
> I don't have a near plan to describe QCA6390 supplies and pins, so don't
> worry. I also don't think Qualcomm BT understand what are bindings, so I
> don't expect patches from them.
> 

For property enable-gpios of QCA6390, it is optional so not marked as
required by the dts spec, dts spec is right, i would like to share
history about QCA6390.

1) it was firstly brought-up by Rocky Liao who is the same team as me.
e5d6468fe9d8 Bluetooth: hci_qca: Add support for Qualcomm Bluetooth SoC
QCA6390

2) for its first product at that time, there are no H/W pin for
enable-gpios, so has issue that BT enable failure after warm reboot. so
i submitted below commit to solve it.
Commit: 7e7bbddd029b ("Bluetooth: hci_qca: Fix qca6390 enable failure
after warm reboot")

3) then Krzysztof Kozlowski submitted below commit to solve
use-after-free issue but also introduced the regression issue
Commit: 272970be3dab ("Bluetooth: hci_qca: Fix driver shutdown on closed
serdev")

4) the issue is reported by Wren recently by below link, and i was
notified to notice this issue and co-work with you to solve it.
https://lore.kernel.org/linux-bluetooth/ea20bb9b-6b60-47fc-ae42-5eed918ad7b4@quicinc.com/T/#m73d6a71d2f454bb03588c66f3ef7912274d37c6f
https://bugzilla.kernel.org/show_bug.cgi?id=218726

5) i known the root cause when i saw the issue description and have
below formal fix [v7,2/2] and it is verified several times and can
solve the issue reported. and the first debugging change included the
fix logic.
https://patchwork.kernel.org/project/bluetooth/list/?series=847290

6) it does not matter if my fix is not expected, please wait for other
good fix.

7) i really don't want to discuss any more since i really talk two much
as shown by below link, i will disappear for a short of time.
https://lore.kernel.org/linux-bluetooth/1713771497-5733-1-git-send-email-quic_zijuhu@quicinc.com/#t

> Best regards,
> Krzysztof
>