Date:   Mon, 30 Jan 2023 14:29:28 +0200
From:   Andy Shevchenko <andriy.shevchenko@linux.intel.com>
To:     Angel Iglesias <ang.iglesiasg@gmail.com>
Cc:     linux-iio@vger.kernel.org, Jonathan Cameron <jic23@kernel.org>,
        Lars-Peter Clausen <lars@metafoo.de>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Andreas Klinger <ak@it-klinger.de>, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 7/7] iio: pressure: bmp280: Add nvmem operations for
 BMP580
Message-ID: <Y9e4KFhB2E6E4EQb@smile.fi.intel.com>
References: <cover.1674954271.git.ang.iglesiasg@gmail.com>
 <013686653a60092227c842ae2fd0197f2509ad7c.1674954271.git.ang.iglesiasg@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <013686653a60092227c842ae2fd0197f2509ad7c.1674954271.git.ang.iglesiasg@gmail.com>
Organization: Intel Finland Oy - BIC 0357606-4 - Westendinkatu 7, 02160 Espoo
Precedence: bulk

On Sun, Jan 29, 2023 at 02:33:09AM +0100, Angel Iglesias wrote:
> The pressure sensor BMP580 contains a non-volatile memory that stores
> trimming and configuration params. That memory provides an programmable
> user range of three 2-byte words.

...

> +#define NVM_READ	false
> +#define NVM_WRITE	true

How is it helpful and why it's not namespaced properly (can collide with
NVM framework)?

...

> +	/* Wait until NVM is ready again */
> +	do {
> +		ret = regmap_read(data->regmap, BMP580_REG_STATUS, &reg);
> +		if (ret) {
> +			dev_err(data->dev, "failed to check nvm status\n");
> +			reg &= ~BMP580_STATUS_NVM_RDY_MASK;
> +		}
> +	} while (time_before(jiffies, deadline) && !(reg & BMP580_STATUS_NVM_RDY_MASK));

regmap_read_poll_timeout()?

> +	if (!(reg & BMP580_STATUS_NVM_RDY_MASK)) {

	if (ret) {
		...
		return ret;
	}

> +		dev_err(data->dev,
> +			"reached timeout waiting for nvm operation completion\n");
> +		return -ETIMEDOUT;
> +	}

...

> +	while (bytes >= sizeof(u16)) {

sizeof(*dst) ?

Or sizeof(data->le16)?

> +		addr = bmp580_nvmem_addrs[offset / sizeof(u16)];

Ditto.

...

> +		bytes -= sizeof(u16);
> +		offset += sizeof(u16);

Ditto.

> +	}

...

> +static int bmp580_nvmem_write(void *priv, unsigned int offset, void *val,
> +			      size_t bytes)
> +{

Same comments as per above function.

>  }


-- 
With Best Regards,
Andy Shevchenko