Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date:   Tue, 09 Apr 2019 22:48:57 +0800
From:   Rocky Liao <rjliao@codeaurora.org>
To:     Rob Herring <robh+dt@kernel.org>
Cc:     Marc Gonzalez <marc.w.gonzalez@free.fr>,
        Marcel Holtmann <marcel@holtmann.org>,
        Johan Hedberg <johan.hedberg@gmail.com>,
        Matthias Kaehlcke <mka@chromium.org>,
        LKML <linux-kernel@vger.kernel.org>,
        "open list:BLUETOOTH DRIVERS" <linux-bluetooth@vger.kernel.org>,
        MSM <linux-arm-msm@vger.kernel.org>,
        Balakrishna Godavarthi <bgodavar@codeaurora.org>
Subject: Re: [PATCH v2 2/2] dt-bindings: net: bluetooth: Add device property
 nvm-postfix for QCA6174
In-Reply-To: <CAL_JsqKLL8x63wzpi2tfNfvVe5TX=Uxz+owPZbhevhW-84bDAA@mail.gmail.com>
References: <1554359858-21049-1-git-send-email-rjliao@codeaurora.org>
 <1554368908-22017-1-git-send-email-rjliao@codeaurora.org>
 <1554368908-22017-2-git-send-email-rjliao@codeaurora.org>
 <535c74d4-00b1-eef2-578a-8807f4b9f640@free.fr>
 <319cbb5f23ed3d9f1ae77c697001378d@codeaurora.org>
 <CAL_JsqKLL8x63wzpi2tfNfvVe5TX=Uxz+owPZbhevhW-84bDAA@mail.gmail.com>
Message-ID: <b7afb25fe2ce759281f1a887df95e71f@codeaurora.org>
User-Agent: Roundcube Webmail/1.2.5
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2019-04-09 22:03, Rob Herring wrote:
> On Tue, Apr 9, 2019 at 5:15 AM Rocky Liao <rjliao@codeaurora.org> 
> wrote:
>> 
>> On 2019-04-04 20:32, Marc Gonzalez wrote:
>> > +robh
>> >
>> > On 04/04/2019 11:08, Rocky Liao wrote:
>> >
>> >> This patchs patch adds an optional device property nvm-postfix to
>> >> allow the
>> >> driver to load customized nvm file based on this property
>> >
>> > While text /before/ is indeed called a "prefix", text /after/ is not a
>> > "postfix",
>> > but a "suffix".
>> >
>> >
>> >>  Documentation/devicetree/bindings/net/qualcomm-bluetooth.txt | 2 ++
>> >>  1 file changed, 2 insertions(+)
>> >>
>> >> diff --git
>> >> a/Documentation/devicetree/bindings/net/qualcomm-bluetooth.txt
>> >> b/Documentation/devicetree/bindings/net/qualcomm-bluetooth.txt
>> >> index 824c0e2..70cda4b 100644
>> >> --- a/Documentation/devicetree/bindings/net/qualcomm-bluetooth.txt
>> >> +++ b/Documentation/devicetree/bindings/net/qualcomm-bluetooth.txt
>> >> @@ -16,6 +16,7 @@ Optional properties for compatible string
>> >> qcom,qca6174-bt:
>> >>
>> >>   - enable-gpios: gpio specifier used to enable chip
>> >>   - clocks: clock provided to the controller (SUSCLK_32KHZ)
>> >> + - nvm-postfix: nvm file postfix to load customized nvm file
>> >
>> > The device tree is supposed to describe hardware.
>> >
>> > The name of which file to load can hardly be considered part of the HW.
>> >
>> > Possible solutions:
>> > 1) derive the file name from the compatible string
>> > 2) pass the name as a module parameter
>> > 3) something else
>> >
>> >
>> >> @@ -39,6 +40,7 @@ serial@7570000 {
>> >>
>> >>              enable-gpios = <&pm8994_gpios 19 GPIO_ACTIVE_HIGH>;
>> >>              clocks = <&divclk4>;
>> >> +            nvm-postfix = "i2s";
>> >>      };
>> >>  };
>> >
>> > If one provides the entire suffix, including the underscore, then you
>> > can
>> > simplify the code:
>> >
>> >       snprintf(config.fwname, sizeof(config.fwname), "qca/nvm_%08x%s.bin",
>> > soc_ver, suffix ? suffix : "");
>> >
>> > Regards
>> .
>> Hi Marc,
>> 
>> The major purpose for that property is about the BT audio bus type, 
>> can
>> it be considered as part of the HW? If yes maybe we can use a property
>> name "audio-bus" to reflect that.
>> 
>> If not then I will adopt the solution 1 to add a new compatible string
>> "{ .compatible = "qcom,qca6174-bt-i2s" }" and load specific nvm for 
>> this
>> compatible string, please feel free to let me know if any other
>> concerns.
> 
> I don't think the suggestion was to add the nvm string to the
> compatible, but rather compatible strings serve as a map key. Having
> board specific firmware files for wifi/bt is pretty common, but
> parameters for 'i2s' is a bit strange. So a better explanation of what
> parameters this contains would help. How/when does it vary, for
> example?
> 
> Also, if it is only a handful of parameters, making them DT properties
> is preferred.
> 
> Rob

Ok, I prefer to go with adding a device property "firmware-name" with 
full firmware name as you suggested in previous comment.
--
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora 
Forum,
a Linux Foundation Collaborative Project