Re: [PATCH] saner warnings for restricted types

[Al Viro <viro@ZenIV.linux.org.uk>]

On Tue, Jan 15, 2008 at 12:36:12PM -0500, Pavel Roskin wrote:
> On Tue, 2008-01-15 at 16:52 +0000, Al Viro wrote:
> >  	if (class & TYPE_RESTRICT) {
> > -		warning(expr->pos, "restricted degrades to integer");
> >  		if (class & TYPE_FOULED)
> >  			*ctype = unfoul(*ctype);
> > +		warning(expr->pos, "%sdegrades to integer",
> > +			show_typename(*ctype));
> 
> Missing space after "%s"?

No.  Try it and you'll see...  The way show_typename() (actually, the
stuff behind it) works you get a space after the damn thing.  What you'll
get is something like

kernel/power/main.c:242:2: warning: restricted suspend_state_t degrades to integer

Actually, if you look at the things like

fs/afs/fsclient.c:465:21: warning: incorrect type in assignment (different base types)
fs/afs/fsclient.c:465:21:    expected restricted __be32 [usertype] operation_ID
fs/afs/fsclient.c:465:21:    got int

you'll see that the last line is "fs/afs/fsclient.c:465:21:    got int \n" -
with whitespace in the end.  We could get rid of that, but it'll take more
massage and it's definitely a separate patch series.