[RFC] ring: further performance improvements with C11
Honnappa Nagarahalli
Honnappa.Nagarahalli at arm.com
Thu Aug 3 04:56:34 CEST 2023
> -----Original Message-----
> From: Thomas Monjalon <thomas at monjalon.net>
> Sent: Monday, July 31, 2023 7:31 AM
> To: Wathsala Wathawana Vithanage <wathsala.vithanage at arm.com>
> Cc: Honnappa Nagarahalli <Honnappa.Nagarahalli at arm.com>;
> konstantin.v.ananyev at yandex.ru; Ruifeng Wang <Ruifeng.Wang at arm.com>;
> dev at dpdk.org; nd <nd at arm.com>; stephen at networkplumber.org;
> jerinj at marvell.com; Morten Brørup <mb at smartsharesystems.com>; Tyler
> Retzlaff <roretzla at linux.microsoft.com>
> Subject: Re: [RFC] ring: further performance improvements with C11
>
> 15/06/2023 22:13, Wathsala Vithanage:
> > For improved performance over the current C11 based ring
> > implementation following changes were made.
> > (1) Replace tail store with RELEASE semantics in
> > __rte_ring_update_tail with a RELEASE fence. Replace load of the tail
> > with ACQUIRE semantics in __rte_ring_move_prod_head and
> > __rte_ring_move_cons_head with ACQUIRE fences.
> > (2) Remove ACQUIRE fences between load of the old_head and load of the
> > cons_tail in __rte_ring_move_prod_head and __rte_ring_move_cons_head.
> > These two fences are not required for the safety of the ring library.
> >
> > Signed-off-by: Wathsala Vithanage <wathsala.vithanage at arm.com>
> > Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli at arm.com>
> > Reviewed-by: Ruifeng Wang <ruifeng.wang at arm.com>
>
> Are we waiting for more reviews?
We do not have a good solution. We should discuss this in the Techboard meeting.
>
More information about the dev
mailing list