Subject: Re: [PATCHv7 3/7] eeprom: at24: extend driver to plug into the NVMEM
 framework
To: Andrew Lunn <andrew@lunn.ch>
References: <1456516764-1456-1-git-send-email-andrew@lunn.ch>
 <1456516764-1456-4-git-send-email-andrew@lunn.ch>
 <56D75F3F.5000806@mleia.com> <20160302214859.GA15541@lunn.ch>
Cc: GregKH <greg@kroah.com>, srinivas.kandagatla@linaro.org,
        maxime.ripard@free-electrons.com, wsa@the-dreams.de,
        broonie@kernel.org, linux-kernel@vger.kernel.org,
        pantelis.antoniou@konsulko.com, bgolaszewski@baylibre.com
From: Vladimir Zapolskiy <vz@mleia.com>
Message-ID: <56D77129.6090100@mleia.com>
Date: Thu, 3 Mar 2016 01:03:05 +0200
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:38.0) Gecko/20100101
 Icedove/38.5.0
MIME-Version: 1.0
In-Reply-To: <20160302214859.GA15541@lunn.ch>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1424
Lines: 58

On 02.03.2016 23:48, Andrew Lunn wrote:
> On Wed, Mar 02, 2016 at 11:46:39PM +0200, Vladimir Zapolskiy wrote:
>> Hi Andrew,
>>
>> On 26.02.2016 21:59, Andrew Lunn wrote:
>>> Add a regmap for accessing the EEPROM, and then use that with the
>>> NVMEM framework. Set the NVMEM config structure to enable backward, so
>>> that the 'eeprom' file in sys is provided by the framework.
>>>
>>> Signed-off-by: Andrew Lunn <andrew@lunn.ch>
>>> Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
>>> Tested-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>>> ---
>>
>> [snip]
>>
>>> +static int at24_regmap_read(void *context, const void *reg, size_t reg_size,
>>> +			    void *val, size_t val_size)
>>> +{
>>> +	struct at24_data *at24 = context;
>>> +	off_t offset = *(u32 *)reg;
>>> +	int err;
>>> +
>>> +	err = at24_read(at24, val, offset, val_size);
>>> +	if (err)
>>> +		return err;
>>> +	return 0;
>>
>> return at24_read(at24, val, offset, val_size);
>>
>> Minus 5 LoC.
> 
> And everything breaks :-(
> 
> regmap expects either an error code, or 0. Return a positive value and
> it is not happy.
> 

Well, do you agree that semantically my proposed change is equal to the
original one?

Let see...

	static int at24_regmap_read() {

		int err;

		err = at24_read(at24, val, offset, val_size);
		if (err)
			return err;
		return 0;
	}

I don't see a check for (err <= 0) returned.

--
With best wishes,
Vladimir