On Tue, May 18, 2021 at 2:57 PM Feifei Wang <feifei.wa...@arm.com> wrote:
>
> Hi, Jerin
>
> Sorry to disturb you. Would you please help me review this patch, thanks very
> much.
In slowpath, I thought of having this barrier. But I agree that it can
be removed.
Since this is not bug. We will merge this patch in next release.
>
>
> Best Regards
> Feifei
>
> > -----Original Message-----
> > From: Feifei Wang <feifei.wa...@arm.com>
> > Sent: Monday, May 10, 2021 2:12 PM
> > To: jer...@marvell.com
> > Cc: dev@dpdk.org; nd <n...@arm.com>; Feifei Wang
> > <feifei.wa...@arm.com>; Ruifeng Wang <ruifeng.w...@arm.com>;
> > Honnappa Nagarahalli <honnappa.nagaraha...@arm.com>
> > Subject: [PATCH] app/eventdev: remove unnecessary barrier for order test
> >
> > For "order_launch_lcores" function, wmb after that the main lcore updates
> > the variable "t->err", which represents the end of the test signal, is
> > unnecessary. Because after the main lcore updates this siginal variable, it
> > will
> > jump out of the launch function loop, and wait other lcores stop or return
> > error in the main function(evt_main.c).
> > During this time, there is no storing operation and thus no need for wmb.
> >
> > Signed-off-by: Feifei Wang <feifei.wa...@arm.com>
> > Reviewed-by: Ruifeng Wang <ruifeng.w...@arm.com>
> > Reviewed-by: Honnappa Nagarahalli <honnappa.nagaraha...@arm.com>
> > ---
> > app/test-eventdev/test_order_common.c | 1 -
> > 1 file changed, 1 deletion(-)
> >
> > diff --git a/app/test-eventdev/test_order_common.c b/app/test-
> > eventdev/test_order_common.c
> > index 04456d56db..d7760061ba 100644
> > --- a/app/test-eventdev/test_order_common.c
> > +++ b/app/test-eventdev/test_order_common.c
> > @@ -308,7 +308,6 @@ order_launch_lcores(struct evt_test *test, struct
> > evt_options *opt,
> > rte_event_dev_dump(opt->dev_id, stdout);
> > evt_err("No schedules for seconds,
> > deadlock");
> > t->err = true;
> > - rte_smp_wmb();
> > break;
> > }
> > old_remaining = remaining;
> > --
> > 2.25.1
> >
>
