On Sat, 26 Jun 2010 17:14:59 +0100 (BST), Maciej W. Rozycki wrote:
> On Tue, 22 Jun 2010, Guenter Roeck wrote:
>
> > > > @@ -104,7 +104,7 @@ static int smbus_xfer(struct i2c_adapter *i2c_adap, u16 addr,
> > > > if (error & M_SMB_ERROR) {
> > > > /* Clear error bit by writing a 1 */
> > > > csr_out32(M_SMB_ERROR, SMB_CSR(adap, R_SMB_STATUS));
> > > > - return -1; /* XXXKW better error code? */
> > > > + return -ENXIO;
> > > > }
> > > >
> > > > if (data_bytes == 1)
> > >
> > > Definitely an improvement. However, returning -ENXIO on all errors
> > > seems wrong. This error value should only be returned on missing ack
> > > from the slave on address byte. Isn't it possible to distinguish
> > > between different error kinds? M_SMB_ERROR_TYPE seems promising, but
> > > one would need to look up the datasheet (which I don't have) to
> > > discover its meaning.
> > >
> > Makes sense. I'll dig up a copy of the datasheet and see if I can improve it.
>
> The M_SMB_ERROR_TYPE bit is cleared if an expected ackonwledgement has
> not been seen and set if the transfer has failed after 15 retries. So
> that's probably ENXIO for the former and EIO for the latter.
Exactly what Guenter did. Good.
--
Jean Delvare
--
To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
