From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ilya Matveychikov <matvejchikov@gmail.com>
Subject: Re: [PATCH v4] mbuf: fix mbuf free performance with non
	atomic refcnt
Date: Fri, 8 Dec 2017 20:04:50 +0400
Message-ID: <60A7E7E9-91D7-48C0-BE51-D284318FD659@gmail.com>
References: <20171115091413.27119-1-hhaim@cisco.com>
 <20171208154651.16546-1-olivier.matz@6wind.com>
Mime-Version: 1.0 (Mac OS X Mail 11.1 \(3445.4.7\))
Content-Type: text/plain;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable
Cc: dev@dpdk.org, "Hanoch Haim (hhaim)" <hhaim@cisco.com>,
 konstantin.ananyev@intel.com
To: Olivier Matz <olivier.matz@6wind.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mail-wm0-f68.google.com (mail-wm0-f68.google.com [74.125.82.68])
 by dpdk.org (Postfix) with ESMTP id F12A3199B6
 for <dev@dpdk.org>; Fri,  8 Dec 2017 17:04:53 +0100 (CET)
Received: by mail-wm0-f68.google.com with SMTP id f9so3976332wmh.0
 for <dev@dpdk.org>; Fri, 08 Dec 2017 08:04:53 -0800 (PST)
In-Reply-To: <20171208154651.16546-1-olivier.matz@6wind.com>
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

Olivier,

> On Dec 8, 2017, at 7:46 PM, Olivier Matz <olivier.matz@6wind.com> =
wrote:
>=20

>=20
> lib/librte_mbuf/rte_mbuf.h | 23 ++++++++++++++++++-----
> 1 file changed, 18 insertions(+), 5 deletions(-)
>=20
> diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
> index ce8a05ddf..dd08cb72b 100644
> --- a/lib/librte_mbuf/rte_mbuf.h
> +++ b/lib/librte_mbuf/rte_mbuf.h
> @@ -764,6 +764,13 @@ rte_mbuf_refcnt_set(struct rte_mbuf *m, uint16_t =
new_value)
> 	rte_atomic16_set(&m->refcnt_atomic, new_value);
> }
>=20
> +/* internal */
> +static inline uint16_t
> +__rte_mbuf_refcnt_update(struct rte_mbuf *m, int16_t value)
> +{
> +	return (uint16_t)(rte_atomic16_add_return(&m->refcnt_atomic, =
value));

What=E2=80=99s the purpose of using direct cast to uint16_t here and in =
other places?

> +}
> +
> /**
>  * Adds given value to an mbuf's refcnt and returns its new value.
>  * @param m
> @@ -788,19 +795,26 @@ rte_mbuf_refcnt_update(struct rte_mbuf *m, =
int16_t value)
> 		return 1 + value;
> 	}
>=20
> -	return (uint16_t)(rte_atomic16_add_return(&m->refcnt_atomic, =
value));
> +	return __rte_mbuf_refcnt_update(m, value);
> }
>=20
> #else /* ! RTE_MBUF_REFCNT_ATOMIC */
>=20
> +/* internal */
> +static inline uint16_t
> +__rte_mbuf_refcnt_update(struct rte_mbuf *m, int16_t value)
> +{
> +	m->refcnt =3D (uint16_t)(m->refcnt + value);
> +	return m->refcnt;
> +}
> +
> /**
>  * Adds given value to an mbuf's refcnt and returns its new value.
>  */
> static inline uint16_t
> rte_mbuf_refcnt_update(struct rte_mbuf *m, int16_t value)
> {
> -	m->refcnt =3D (uint16_t)(m->refcnt + value);
> -	return m->refcnt;
> +	return __rte_mbuf_refcnt_update(m, value);
> }
>=20
> /**
> @@ -1364,8 +1378,7 @@ rte_pktmbuf_prefree_seg(struct rte_mbuf *m)
>=20
> 		return m;
>=20
> -       } else if (rte_atomic16_add_return(&m->refcnt_atomic, -1) =3D=3D=
 0) {
> -
> +	} else if (__rte_mbuf_refcnt_update(m, -1) =3D=3D 0) {
>=20
> 		if (RTE_MBUF_INDIRECT(m))
> 			rte_pktmbuf_detach(m);
> --=20
> 2.11.0
>=20