Re: [PATCH 1/3] evaluate: warn on identical exprs around '&&'

[Josh Triplett <josh@xxxxxxxxxxxxxxxx>]

On Mon, Aug 29, 2011 at 12:01:06PM +0200, Jonathan Neuschäfer wrote:
> On Sun, Aug 28, 2011 at 03:14:18PM +1200, Chris Forbes wrote:
> > +	case EXPR_BINOP:
> > +	case EXPR_COMMA:
> > +	case EXPR_COMPARE:
> > +	case EXPR_LOGICAL:
> > +	case EXPR_ASSIGNMENT:
> > +		return expr_equiv(lhs->left, rhs->left) &&
> > +			expr_equiv(lhs->right, rhs->right);
> [...]
> > +	if ((a == b) && (a == b))	/* should warn */
> > +		bar();
> > +
> > +	if ((a == b) && (b == c))	/* should not warn */
> > +		bar();
> 
> Should it maybe also handle cases like this?
> 
> if ((a == b) && (b == a))
> 	bar();

That seems both significantly harder to handle and significantly less
likely (since it won't occur due to simple copy/paste).  Probably worth
doing at some point, but I don't think a first pass needs to consider
it.

- Josh Triplett
--
To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html