[RFC PATCH] Make "bad constant expression" a warning, not an error

[RFC PATCH] Make "bad constant expression" a warning, not an error

[Pavel Roskin]

Do the same to "bad integer constant expression".  It's still possible
to continue parsing the code if those are not constant.  To be on the
safe side, assume the actual value to be the largest value for the
type.
---

 expand.c |   12 +++++++-----
 1 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/expand.c b/expand.c
index 032f0c5..12a8989 100644
--- a/expand.c
+++ b/expand.c
@@ -1201,18 +1201,20 @@ static long long __get_expression_value(struct expression *expr, int strict)
 		expression_error(expr, "bad constant expression type");
 		return 0;
 	}
+
+	mask = 1ULL << (ctype->bit_size-1);
+
 	expand_expression(expr);
 	if (expr->type != EXPR_VALUE) {
-		expression_error(expr, "bad constant expression");
-		return 0;
+		warning(expr->pos, "bad constant expression");
+		return mask;
 	}
 	if (strict && bad_integer_constant_expression(expr)) {
-		expression_error(expr, "bad integer constant expression");
-		return 0;
+		warning(expr->pos, "bad integer constant expression");
+		return mask;
 	}
 
 	value = expr->value;
-	mask = 1ULL << (ctype->bit_size-1);
 
 	if (value & mask) {
 		while (ctype->type != SYM_BASETYPE)

-- 
Regards,
Pavel Roskin

Re: [RFC PATCH] Make "bad constant expression" a warning, not an error

[Christopher Li]

I object.

You should fix it in the parser side. When the parser call get expression
value, it really expect to get back a constant value. Make it return the
largest value on error is just wrong.

Chris

On Fri, Jun 20, 2008 at 8:08 AM, Pavel Roskin <proski@gnu.org> wrote:
> Do the same to "bad integer constant expression".  It's still possible
> to continue parsing the code if those are not constant.  To be on the
> safe side, assume the actual value to be the largest value for the
> type.
> ---
>
>  expand.c |   12 +++++++-----
>  1 files changed, 7 insertions(+), 5 deletions(-)
>
> diff --git a/expand.c b/expand.c
> index 032f0c5..12a8989 100644
> --- a/expand.c
> +++ b/expand.c
> @@ -1201,18 +1201,20 @@ static long long __get_expression_value(struct expression *expr, int strict)
>                expression_error(expr, "bad constant expression type");
>                return 0;
>        }
> +
> +       mask = 1ULL << (ctype->bit_size-1);
> +
>        expand_expression(expr);
>        if (expr->type != EXPR_VALUE) {
> -               expression_error(expr, "bad constant expression");
> -               return 0;
> +               warning(expr->pos, "bad constant expression");
> +               return mask;
>        }
>        if (strict && bad_integer_constant_expression(expr)) {
> -               expression_error(expr, "bad integer constant expression");
> -               return 0;
> +               warning(expr->pos, "bad integer constant expression");
> +               return mask;
>        }
>
>        value = expr->value;
> -       mask = 1ULL << (ctype->bit_size-1);
>
>        if (value & mask) {
>                while (ctype->type != SYM_BASETYPE)
>
> --
> Regards,
> Pavel Roskin
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>