Re: [PATCH] i2c/busses: (mpc) Add support for SMBUS_READ_BLOCK_DATA

[Jean Delvare <khali@linux-fr.org>]

On Wed, 16 Nov 2011 09:28:29 -0800, York Sun wrote:
> I am not sure if this mail will reach the list as I am not subscribed.
> 
> In the v2 patch (actually both version), you wrote
> 
> > +		byte = readb(i2c->base + MPC_I2C_DR);
> > +
> > +		/* Adjust length if first received byte is length */
> > +		if (i == 0 && recv_len) {
> > +			if (byte == 0 || byte > I2C_SMBUS_BLOCK_MAX)
> > +				return -EPROTO;
> > +			length += byte;
> > +			/*
> > +			 * For block reads, generate txack here if data length
> > +			 * is 1 byte (total length is 2 bytes).
> > +			 */
> > +			if (length == 2)
> > +				writeccr(i2c, CCR_MIEN | CCR_MEN | CCR_MSTA
> > +					 | CCR_TXAK);
> > +		}
> > +		data[i] = byte;
> 
> I am wondering if the first byte should be left out for data[] if the
> recv_len is non-zero. If I understand correctly, this patch is to
> support of SMBus with multiple byte read. The SMBus is different from
> I2C bus on multiple byte. The first data is byte count vs slave data for
> I2C. So you will receive all data preceded by the byte count, which is
> one more than your loop.

That's why he wrote

	length += byte;

and not

	length = byte;

length is initialized to 1 for SMBus block reads.

-- 
Jean Delvare