On Tue, Jul 12, 2016 at 09:15:40AM +0300, Liran Schour wrote:
> Hashing will not be consistent as long as we use for hashing the index of the
> bucket in the list (for remove/insert of buckets not from/to the end of the
> bucket list).
> Use bucket_id for hashing instead.
>
> Signed-off-by: Liran Schour <lir...@il.ibm.com>
Acked-by: Simon Horman <simon.hor...@netronome.com>
> ---
> ofproto/ofproto-dpif-xlate.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/ofproto/ofproto-dpif-xlate.c b/ofproto/ofproto-dpif-xlate.c
> index 1977b6b..655033d 100644
> --- a/ofproto/ofproto-dpif-xlate.c
> +++ b/ofproto/ofproto-dpif-xlate.c
> @@ -1558,7 +1558,6 @@ group_best_live_bucket(const struct xlate_ctx *ctx,
> {
> struct ofputil_bucket *best_bucket = NULL;
> uint32_t best_score = 0;
> - int i = 0;
>
> struct ofputil_bucket *bucket;
> const struct ovs_list *buckets;
> @@ -1566,13 +1565,13 @@ group_best_live_bucket(const struct xlate_ctx *ctx,
> group_dpif_get_buckets(group, &buckets);
> LIST_FOR_EACH (bucket, list_node, buckets) {
> if (bucket_is_alive(ctx, bucket, 0)) {
> - uint32_t score = (hash_int(i, basis) & 0xffff) * bucket->weight;
> + uint32_t score =
> + (hash_int(bucket->bucket_id, basis) & 0xffff) *
> bucket->weight;
> if (score >= best_score) {
> best_bucket = bucket;
> best_score = score;
> }
> }
> - i++;
> }
>
> return best_bucket;
> --
> 2.1.4
>
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
