Re: [PATCH] watchdog: pcwd_pci: mark expected switch fall-through

[Guenter Roeck]

On Fri, Nov 03, 2017 at 09:02:33AM -0700, Joe Perches wrote:
> On Fri, 2017-11-03 at 08:54 -0700, Guenter Roeck wrote:
> > On Fri, Nov 03, 2017 at 04:04:23PM +0100, Wim Van Sebroeck wrote:
> > > Hi Gustavo,
> > > 
> > > > In preparation to enabling -Wimplicit-fallthrough, mark switch cases
> > > > where we are expecting to fall through.
> > > > 
> > > > Notice that in this particular case I replaced "Fall" with a proper
> > > > "fall through" comment, which is what GCC is expecting to find.
> > > > 
> > > > Signed-off-by: Gustavo A. R. Silva <garsilva@xxxxxxxxxxxxxx>
> > > > ---
> > > >  drivers/watchdog/pcwd_pci.c | 2 +-
> > > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > > 
> > > > diff --git a/drivers/watchdog/pcwd_pci.c b/drivers/watchdog/pcwd_pci.c
> > > > index c0d07ee..c882252 100644
> > > > --- a/drivers/watchdog/pcwd_pci.c
> > > > +++ b/drivers/watchdog/pcwd_pci.c
> > > > @@ -545,7 +545,7 @@ static long pcipcwd_ioctl(struct file *file, unsigned int cmd,
> > > >  			return -EINVAL;
> > > >  
> > > >  		pcipcwd_keepalive();
> > > > -		/* Fall */
> > > > +		/* fall through */
> > > >  	}
> > > >  
> > > >  	case WDIOC_GETTIMEOUT:
> > > > -- 
> > > > 2.7.4
> > > > 
> > > 
> > > Shouldn't the /* fall through */ come after the } ?
> > > 
> > 
> > Good question. This is an unconditional code block needed to declare
> > a local variable within the case statement. What is correct in that
> > situation ?
> 
> I think it'd be clearer to avoid the trivial fallthrough
> optimization/complexity and just directly use
> 
> 		return put_user(new_heartbeat, p);
> 
> as heartbeat and new_heartbeat are now the same value here.
> 
I don't think it really matters. What would matter would be for someone
to convert he driver to use the watchdog subsystem.

Guenter