On Tue, Oct 17, 2017 at 04:41:44PM +0530, Rao, Nikhil wrote:
<snip>
> > static __rte_always_inline uint16_t
> > ssows_get_work(struct ssows *ws, struct rte_event *ev)
> > {
> >@@ -62,9 +86,10 @@ ssows_get_work(struct ssows *ws, struct rte_event *ev)
> > ws->cur_tt = sched_type_queue & 0x3;
> > ws->cur_grp = sched_type_queue >> 2;
> > sched_type_queue = sched_type_queue << 38;
> >-
> > ev->event = sched_type_queue | (get_work0 & 0xffffffff);
> >- ev->u64 = get_work1;
> >+ ev->u64 = get_work1 && !ev->event_type ?
>
> I think the code readability would be better if you compared event_type to
> RTE_EVENT_TYPE_ETHDEV.
>
Agreed,
Will send out a v2.
Thanks,
Pavan.
> Reviewed by: Nikhil Rao <nikhil....@intel.com>
