Re: [PATCH] iio: accel: cros_ec_accel_legacy: Mark expected switch fall-throughs

[Marcus Folkesson]

Hi Gutavo,

Sorry for the delay.

On Wed, Aug 15, 2018 at 12:50:10PM -0500, Gustavo A. R. Silva wrote:
> Hi Marcus,
> 
> On 8/15/18 12:27 PM, Marcus Folkesson wrote:
> > Hi,
> > 
> > On Wed, Aug 15, 2018 at 11:38:52AM -0500, Gustavo A. R. Silva wrote:
> >> In preparation to enabling -Wimplicit-fallthrough, mark switch cases
> >> where we are expecting to fall through.
> >>
> >> Addresses-Coverity-ID: 1462408 ("Missing break in switch")
> >> Signed-off-by: Gustavo A. R. Silva <gustavo@xxxxxxxxxxxxxx>
> >> ---
> >>  drivers/iio/accel/cros_ec_accel_legacy.c | 2 ++
> >>  1 file changed, 2 insertions(+)
> >>
> >> diff --git a/drivers/iio/accel/cros_ec_accel_legacy.c b/drivers/iio/accel/cros_ec_accel_legacy.c
> >> index 063e89e..d609654 100644
> >> --- a/drivers/iio/accel/cros_ec_accel_legacy.c
> >> +++ b/drivers/iio/accel/cros_ec_accel_legacy.c
> >> @@ -385,8 +385,10 @@ static int cros_ec_accel_legacy_probe(struct platform_device *pdev)
> >>  		switch (i) {
> >>  		case X:
> >>  			ec_accel_channels[X].scan_index = Y;
> >> +			/* fall through */
> >>  		case Y:
> >>  			ec_accel_channels[Y].scan_index = X;
> >> +			/* fall through */
> >>  		case Z:
> >>  			ec_accel_channels[Z].scan_index = Z;
> >>  		}
> > 
> > Hum, I'm not sure we are supposed to fall through here, even if it does
> > not hurt to do so.
> 
> Yeah. You're right. It doesn't hurt but is actually redundant. I think
> the original intention was to break instead of falling through.
> 
> > I even think we can remove the switch and put that outside the for-loop,
> > e.g:
> > 
> > 	ec_accel_channels[X].scan_index = Y;
> > 	ec_accel_channels[Y].scan_index = X;
> > 	ec_accel_channels[Z].scan_index = Z;
> > 
> > 	for (i = X ; i < MAX_AXIS; i++) {
> > 		if (state->sensor_num == MOTIONSENSE_LOC_LID && i != Y)
> > 			state->sign[i] = -1;
> > 		else
> > 			state->sign[i] = 1;
> > 	}
> > 
> 
> I like this, but the code clearly depends on MAX_AXIS. So, if MAX_AXIS
> will be always 3, then the change you suggest is just fine. Otherwise,
> it seems that adding a break to each case is the right way to go.
> 
> What do you think?

Well, I guess it is a matter of taste after all.
I don't think the number of axis will change, but just put the break in
place is good enough.

Anyway, If we choose to not use the switch, I think we should remove the
for-loop as well, eg:

	ec_accel_channels[X].scan_index = Y;
	ec_accel_channels[Y].scan_index = X;
	ec_accel_channels[Z].scan_index = Z;

	if (state->sensor_num == MOTIONSENSE_LOC_LID) {
		state->sign[X] = -1;
		state->sign[Y] = 1;
		state->sign[Z] = -1;
	} else {
		state->sign[X] = 1;
		state->sign[Y] = 1;
		state->sign[Z] = 1;
	}

But someone else may like to give their point of view on this change.

> 
> Thanks for the feedback.
> --
> Gustavo

Best regards
Marcus Folkesson

Attachment: signature.asc
Description: PGP signature