[PATCH 6/6] [RFC] mtd: mxc-nand: Warn on unimplemented commands

[u.kleine-koenig@pengutronix.de (Uwe Kleine-König)]

Hello,

On Wed, Feb 11, 2015 at 10:40:16AM +0100, Lothar Wa?mann wrote:
> Uwe Kleine-K?nig wrote:
> > On Wed, Feb 11, 2015 at 09:42:56AM +0100, Lothar Wa?mann wrote:
> > > > diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c
> > > > index 0083b4ee4f33..372e0e38f59b 100644
> > > > --- a/drivers/mtd/nand/mxc_nand.c
> > > > +++ b/drivers/mtd/nand/mxc_nand.c
> > > > @@ -1160,6 +1160,10 @@ static void mxc_nand_command(struct mtd_info *mtd, unsigned command,
> > > >  		memcpy32_fromio(host->data_buf, host->main_area0, 512);
> > > >  		host->buf_start = 0;
> > > >  		break;
> > > > +	default:
> > > > +		WARN_ONCE(1, "Unimplemented command (cmd=%u)\n",
> > > > +			  command);
> > > > +		break;
> > > >  	}
> > > useless break;
> > Do you mean the line break? That's right, I fixed it here for a later
> > v2. But I guess you mean the (literal) break here. Right, it could be
> > dropped without change in semantic, but I thought adding it matches the
> > usually recommended style?!
> > 
> Documentation/CodingStyle has this example:
> |        default:
> |                break;
> |        }
> but there is no useful statement in the 'default' case, so the
> 'break' is necessary here.
> IMO this doesn't mandate to add a 'break' at the end of the default
> clause if there are actual statements in this path.
OK, we agree that Documentation/CodingStyle isn't explict about this
case here. For non-default case labels I already saw review requests to
add an explicit break which I consider sensible. For default it's not
that important but IMHO still good style. I want to keep it.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |