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;
Date:   Thu, 6 Dec 2018 15:40:47 +0300
From:   Dan Carpenter <dan.carpenter@oracle.com>
To:     Jeremy Fertic <jeremyfertic@gmail.com>
Cc:     Shreeya Patel <shreeya.patel23498@gmail.com>,
        devel@driverdev.osuosl.org, lars@metafoo.de,
        Michael.Hennerich@analog.com, linux-iio@vger.kernel.org,
        gregkh@linuxfoundation.org, linux-kernel@vger.kernel.org,
        pmeerw@pmeerw.net, knaack.h@gmx.de, jic23@kernel.org
Subject: Re: [PATCH] Revert "Staging: iio: adt7316: Add an extra check for
 'ret' equals to 0"
Message-ID: <20181206124047.GH3095@unbuntlaptop>
References: <20181205014900.4827-1-jeremyfertic@gmail.com>
 <fccdb26936a9cfcd5134c760cd918929ad3dfa19.camel@gmail.com>
 <20181205215953.GA2365@r2700x.localdomain>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20181205215953.GA2365@r2700x.localdomain>
User-Agent: Mutt/1.9.4 (2018-02-28)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Dec 05, 2018 at 02:59:53PM -0700, Jeremy Fertic wrote:
> On Thu, Dec 06, 2018 at 01:25:55AM +0530, Shreeya Patel wrote:
> > On Tue, 2018-12-04 at 18:49 -0700, Jeremy Fertic wrote:
> > > This reverts commit 00426e99789357dbff7e719a092ce36a3ce49d94.
> > > 
> > > i2c_smbus_read_byte() returns 0 when a byte with the value 0 is read
> > > from
> > > the device. This is a valid read so revert the check for 0.
> > > 
> > > Signed-off-by: Jeremy Fertic <jeremyfertic@gmail.com>
> > > ---
> > 
> > Hi Jeremy,
> > 
> > As per my understanding, 0 value indicates no error but no data read.
> > Then how can this be a valid case?
> > 
> > Can you please make me understand that how can we consider this as a
> > valid case even when no data has been read?

It's not reading no data.  It's reading one byte of data and returning
it.

> > 
> > 
> > Thanks
> 
> I'm not sure I understand why the value 0 would indicate no data read.
> Doesn't that just mean a byte was read with the value 0.

Yes.  It does mean that.  Please don't ask rhetorical questions...  :(
This list is full of people who can't resist answering every question.

> For instance, if the input to the adc is 0V. Can you point me to where
> you're seeing that this would indicate no data read?

drivers/i2c/i2c-core-smbus.c
    88  /**
    89   * i2c_smbus_read_byte - SMBus "receive byte" protocol
    90   * @client: Handle to slave device
    91   *
    92   * This executes the SMBus "receive byte" protocol, returning negative errno
    93   * else the byte received from the device.
    94   */
    95  s32 i2c_smbus_read_byte(const struct i2c_client *client)
    96  {
    97          union i2c_smbus_data data;
    98          int status;
    99  
   100          status = i2c_smbus_xfer(client->adapter, client->addr, client->flags,
   101                                  I2C_SMBUS_READ, 0,
   102                                  I2C_SMBUS_BYTE, &data);
   103          return (status < 0) ? status : data.byte;
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   104  }
   105  EXPORT_SYMBOL(i2c_smbus_read_byte);

You are right.  Commit 00426e997893 ("Staging: iio: adt7316: Add an
extra check for 'ret' equals to 0") needs to be reverted...

regards,
dan carpenter