On Tue, Aug 14, 2018 at 05:00:53PM +0200, Norbert Manthey wrote:
> To get more out of the analysis, increase some buffers and limits.
>
> Signed-off-by: Norbert Manthey <nmanthey@xxxxxxxxx>
> ---
> smatch_math.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/smatch_math.c b/smatch_math.c
> index de9f6c0..fef2baf 100644
> --- a/smatch_math.c
> +++ b/smatch_math.c
> @@ -1137,7 +1137,7 @@ static struct range_list *_get_rl(struct expression *expr, int implied, int *rec
> if (!expr)
> return NULL;
>
> - if (++(*recurse_cnt) >= 200)
> + if (++(*recurse_cnt) >= 500)
> return NULL;
Hopefully, this doesn't matter. If it does then probably we should look
at a different way to fix the problem.
>
> switch(expr->type) {
> @@ -1199,7 +1199,7 @@ out_cast:
> struct {
> struct expression *expr;
> struct range_list *rl;
> -} cached_results[24];
> +} cached_results[72];
The idea here would be that this is a speed up. Does it really make a
difference?
regards,
dan carpenter
