On Fri, 11 Mar 2022 06:24:35 +0000
"Ajmera, Megha" <megha.ajm...@intel.com> wrote:
> >
> > > diff --git a/lib/sched/rte_sched.c b/lib/sched/rte_sched.c index
> > > ec74bee939..1d05089d00 100644
> > > --- a/lib/sched/rte_sched.c
> > > +++ b/lib/sched/rte_sched.c
> > > @@ -155,6 +155,7 @@ struct rte_sched_subport {
> > > uint64_t tc_credits[RTE_SCHED_TRAFFIC_CLASSES_PER_PIPE];
> > >
> > > /* TC oversubscription */
> > > + uint8_t is_tc_ov_enabled;
> > > uint64_t tc_ov_wm;
> > > uint64_t tc_ov_wm_min;
> > > uint64_t tc_ov_wm_max;
> >
> > Putting the field there creates a hole in the structure.
> > Put it after tc_ov and fill an existing hole.
> >
> > This is pahole of current code, looks like this struct could use some work
> > to be
> > better packed and aligned.
> >
>
> Thanks Stephen for pointing this out. I agree there is some work needed to
> pack this structure better.
> Can we take this up in a separate patch in later release as it requires more
> performance test runs to see the impact ?
>
Please add new field in existing hole
