Re: [PATCH v2] ASoC: cros_ec_codec: Log results when EC commands fail

[Yu-Hsuan Hsu]

Tzung-Bi Shih <tzungbi@xxxxxxxxxx> æ 2020å7æ3æ éä äå3:32åéï
>
> On Fri, Jul 3, 2020 at 3:19 PM Yu-Hsuan Hsu <yuhsuan@xxxxxxxxxxxx> wrote:
> > Log results of failed EC commands to identify a problem more easily.
> >
> > Replace cros_ec_cmd_xfer_status with cros_ec_cmd_xfer because the result
> > has already been checked in this function. The wrapper is not needed.
>
> Alternatively, you can still use cros_ec_cmd_xfer_status( ).  I guess
> it is okay to have 2 logs for an error.
>
> > diff --git a/sound/soc/codecs/cros_ec_codec.c b/sound/soc/codecs/cros_ec_codec.c
> > index 8d45c628e988e..a4ab62f59efa6 100644
> > --- a/sound/soc/codecs/cros_ec_codec.c
> > +++ b/sound/soc/codecs/cros_ec_codec.c
> > @@ -90,10 +90,17 @@ static int send_ec_host_command(struct cros_ec_device *ec_dev, uint32_t cmd,
> >         if (outsize)
> >                 memcpy(msg->data, out, outsize);
> >
> > -       ret = cros_ec_cmd_xfer_status(ec_dev, msg);
> > +       ret = cros_ec_cmd_xfer(ec_dev, msg);
> >         if (ret < 0)
> I am thinking of if it is a better solution to print msg->result here.
The problem is the msg->result is not always meaningful.
In cros_ec_cmd_xfer_status, we know that the msg->result is meaningful
only when ret == 0. Therefore, we can not print the msg->result
directly here.

In addition, adding a conditional operator here to check whether ret
is -EPROTO is not a good way, either.
We should consider the situation that cros_ec_cmd_xfer may return
-EPROTO directly.

>
> >                 goto error;
> >
> > +       if (msg->result != EC_RES_SUCCESS) {
> > +               dev_err(ec_dev->dev, "Command %d failed: %d\n", cmd,
> > +                       msg->result);
> > +               ret = -EPROTO;
> > +               goto error;
> > +       }
> So that you don't need this block.