Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <94e80a49-11ae-e5b0-7eea-6ed4ec6d2ac8@linaro.org>
Date:   Mon, 26 Jun 2023 15:20:35 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.12.0
From:   Neil Armstrong <neil.armstrong@linaro.org>
Reply-To: neil.armstrong@linaro.org
Subject: Re: [PATCH v3 4/4] input: touchscreen: add SPI support for Goodix
 Berlin Touchscreen IC
Content-Language: en-US
To:     Jeff LaBundy <jeff@labundy.com>
Cc:     Dmitry Torokhov <dmitry.torokhov@gmail.com>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Conor Dooley <conor+dt@kernel.org>,
        Bastien Nocera <hadess@hadess.net>,
        Hans de Goede <hdegoede@redhat.com>,
        Henrik Rydberg <rydberg@bitmath.org>,
        linux-input@vger.kernel.org, linux-arm-msm@vger.kernel.org,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20230606-topic-goodix-berlin-upstream-initial-v3-0-f0577cead709@linaro.org>
 <20230606-topic-goodix-berlin-upstream-initial-v3-4-f0577cead709@linaro.org>
 <ZJiXopmFr4dPbqll@nixie71> <e36a697f-a54d-7bdf-1e18-38f72ec2966c@linaro.org>
 <ZJmMJxXxLrC9Xevi@nixie71>
Organization: Linaro Developer Services
In-Reply-To: <ZJmMJxXxLrC9Xevi@nixie71>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi,

On 26/06/2023 15:01, Jeff LaBundy wrote:
> Hi Neil,
> 
> On Mon, Jun 26, 2023 at 09:02:16AM +0200, Neil Armstrong wrote:
> 
> [...]
> 
>>>> +static int goodix_berlin_spi_probe(struct spi_device *spi)
>>>> +{
>>>> +	struct regmap_config *regmap_config;
>>>> +	struct regmap *regmap;
>>>> +	size_t max_size;
>>>> +	int error = 0;
>>>> +
>>>> +	regmap_config = devm_kmemdup(&spi->dev, &goodix_berlin_spi_regmap_conf,
>>>> +				     sizeof(*regmap_config), GFP_KERNEL);
>>>> +	if (!regmap_config)
>>>> +		return -ENOMEM;
>>>
>>> Is there any reason we cannot simply pass goodix_berlin_spi_regmap_conf to
>>> devm_regmap_init() below? Why to duplicate and pass the copy?
>>>
>>> For reference, BMP280 in IIO is a similar example of a device with regmap
>>> sitting atop a bespoke SPI protocol; it does not seem to take this extra
>>> step.
>>
>> The goodix_berlin_spi_regmap_conf copy is modified after with the correct
>> max raw read/write size, and I'm not a fan of modifying a global structure
>> that could be use for multiple probes, I can make a copy in a stack variable
>> if it feels simpler.
> 
> Ah, that makes sense; in that case, the existing implementation seems fine
> to me. No changes necessary.
> 
> Correct me if I'm wrong, but the stack variable method wouldn't work since
> that memory is gone after goodix_berlin_spi_probe() returns.

The config is only needed for the devm_regmap_init() duration, so keeping
the memory allocated for the whole lifetime of the device seems useless.

Neil

> 
> Kind regards,
> Jeff LaBundy