HI,
On Wed, Jan 23, 2013 at 10:10:31PM +0200, Aaro Koskinen wrote:
> Hi,
>
> Some minor comments/questions:
>
> On Wed, Jan 23, 2013 at 12:23:07PM +0200, Felipe Balbi wrote:
> > @@ -639,31 +630,37 @@ static int omap_i2c_xfer_msg(struct i2c_adapter *adap,
> > dev_err(dev->dev, "controller timed out\n");
> > omap_i2c_reset(dev);
> > __omap_i2c_init(dev);
> > - return -ETIMEDOUT;
> > + ret = -ETIMEDOUT;
> > + goto out;
> > }
> >
> > - if (likely(!dev->cmd_err))
> > - return 0;
> > -
> > /* We have an error */
> > if (dev->cmd_err & (OMAP_I2C_STAT_AL | OMAP_I2C_STAT_ROVR |
> > OMAP_I2C_STAT_XUDF)) {
> > omap_i2c_reset(dev);
> > __omap_i2c_init(dev);
> > - return -EIO;
> > + ret = -EIO;
> > + goto out;
>
> Why change return to goto on these paths? If you reset and re-initialize
> the controller, I think it's unlikely that omap_i2c_wait_for_ardy()
> will ever succeed?
good catch, I'll update.
--
balbi
Attachment:
signature.asc
Description: Digital signature
