From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from EUR02-AM5-obe.outbound.protection.outlook.com (mail-eopbgr00084.outbound.protection.outlook.com [40.107.0.84]) by dpdk.org (Postfix) with ESMTP id 702E51BA30 for ; Wed, 25 Oct 2017 23:42:49 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Mellanox.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version; bh=4KWZp/BvwMl9A7ekWSDwo4HPVE+IV7AG9VIrI7+qKAk=; b=werU0WIy4+Hss4qH41khTmzs2rGgDKAeFMOBvisfXlb8ADs5KiE5ifQFQbtw/mpoL1dY+K/8AIDNY6sVdVaO27LIcxEXHLyvewZhLbMH+M/FdRZobnDn/ZRVKn7Q67IeLKGT0/+TAOIZBHAqKazEYfD3iBBups5bFPK3dlIsrik= Received: from DB5PR05MB1254.eurprd05.prod.outlook.com (10.162.157.140) by DB5PR05MB1254.eurprd05.prod.outlook.com (10.162.157.140) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA384_P256) id 15.20.156.4; Wed, 25 Oct 2017 21:42:46 +0000 Received: from DB5PR05MB1254.eurprd05.prod.outlook.com ([fe80::69a8:7948:e0cb:c16e]) by DB5PR05MB1254.eurprd05.prod.outlook.com ([fe80::69a8:7948:e0cb:c16e%13]) with mapi id 15.20.0156.007; Wed, 25 Oct 2017 21:42:46 +0000 From: Ophir Munk To: Adrien Mazarguil CC: "dev@dpdk.org" , Thomas Monjalon , "Olga Shern" , Matan Azrad Thread-Topic: [PATCH v2 2/7] net/mlx4: inline more Tx functions Thread-Index: AQHTTApqth6x0LV1TkeHYFXL14f/5qL0ymgAgABGhrA= Date: Wed, 25 Oct 2017 21:42:46 +0000 Message-ID: References: <1508752838-30408-1-git-send-email-ophirmu@mellanox.com> <1508768520-4810-1-git-send-email-ophirmu@mellanox.com> <1508768520-4810-3-git-send-email-ophirmu@mellanox.com> <20171025164938.GH26782@6wind.com> In-Reply-To: <20171025164938.GH26782@6wind.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: authentication-results: spf=none (sender IP is ) smtp.mailfrom=ophirmu@mellanox.com; x-originating-ip: [93.173.17.154] x-ms-publictraffictype: Email x-microsoft-exchange-diagnostics: 1; DB5PR05MB1254; 6:XcbuWSnmkNT1kR1lFSYXNAXEPmaJUi2EeMW9ff46e416xDNaBzJtX8fThcBmBrAtKApJGRPjoB8Pr+sbY1l5viyeTL0ZnS5yD2GQ1ajz/JENEhmlpyPPl9NOTy8ghHRubfOFYCOrxHIRbodh1MrWY+QGI2LPicUPTF15Rl9/+rLVBErt/BVmKAqFwfomZ0BZcwPlpIB6ZrA7jgRFms0vOBgi+PbmCDNAQQUD8kSCjgUP0xqeon/fF0ngvcigOQ19P0KismkmY2JwXzFJbA3xE8/y9xhFLdw3cIC0YjMCPJbRpD0lvarhbU23dM1fZPyv2FDd0IvFNDQPiSQ9uuUuYw==; 5:DXzFb0DVTGAnVP7KVKeCbG5MsC8VmbuN6a4ivbM3WqXbemtSMzTKTu1/RwcwsZ23NvyNHaLzHQduMtgb8u6Km5V0O9joQ2+E7lT09V9igdYemDHw3zl/A2ion2+w9akh8t1Xe0hanI3fLenaNcJuNw==; 24:sCxDDpm1rnZCQGjFX65So97moCF2rHJymDyAWvhpJTe0OSp48pcueqEsCNtsnolWWJeWjzx8Lg4c2K68RlTWtgr4c1r+Vftjjm1YSBnFCCE=; 7:4fGXA378ISyzN71qd3eFtvB46rrD7wcqN4u+BfFrPGN8RZQpQK208lW19DEpAsR4F9l1T71udbKIyUB9btFQhv/Ylot0B9VbwTvvzZ2Cv3z97PanMoEmevCmRG6PR3AhkTogYT9Le4TM2jEyXDFN4N4Rv3S5LwwkoR7eB+a8qRFipvZD8I+fjS0p/SCuTk2h+qIMwxwWWojx9ysvT7iukQMUyddCJnKa3zm+MYMKnLw= x-ms-exchange-antispam-srfa-diagnostics: SSOS; x-ms-office365-filtering-correlation-id: af739c1e-b659-4109-50a7-08d51bf154e4 x-ms-office365-filtering-ht: Tenant x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:(22001)(48565401081)(4534020)(4602075)(4627075)(201703031133081)(201702281549075)(2017052603199); SRVR:DB5PR05MB1254; x-ms-traffictypediagnostic: DB5PR05MB1254: x-ld-processed: a652971c-7d2e-4d9b-a6a4-d149256f461b,ExtAddr x-exchange-antispam-report-test: UriScan:(17755550239193); x-microsoft-antispam-prvs: x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(100000700101)(100105000095)(100000701101)(100105300095)(100000702101)(100105100095)(6040450)(2401047)(8121501046)(5005006)(93006095)(93001095)(10201501046)(100000703101)(100105400095)(3231020)(3002001)(6055026)(6041248)(20161123558100)(20161123562025)(20161123555025)(20161123564025)(201703131423075)(201702281528075)(201703061421075)(201703061406153)(20161123560025)(6072148)(201708071742011)(100000704101)(100105200095)(100000705101)(100105500095); SRVR:DB5PR05MB1254; BCL:0; PCL:0; RULEID:(100000800101)(100110000095)(100000801101)(100110300095)(100000802101)(100110100095)(100000803101)(100110400095)(100000804101)(100110200095)(100000805101)(100110500095); SRVR:DB5PR05MB1254; x-forefront-prvs: 0471B73328 x-forefront-antispam-report: SFV:NSPM; SFS:(10009020)(6009001)(376002)(346002)(39860400002)(189002)(199003)(24454002)(25786009)(93886005)(74316002)(5250100002)(478600001)(106356001)(105586002)(229853002)(86362001)(33656002)(7696004)(6246003)(102836003)(7736002)(3846002)(107886003)(2900100001)(53546010)(81166006)(55016002)(99286003)(8676002)(6506006)(81156014)(50986999)(305945005)(54356999)(3660700001)(6116002)(66066001)(54906003)(76176999)(9686003)(4326008)(3280700002)(101416001)(316002)(68736007)(53936002)(6436002)(2906002)(6916009)(189998001)(14454004)(8936002)(97736004)(2950100002)(5660300001); DIR:OUT; SFP:1101; SCL:1; SRVR:DB5PR05MB1254; H:DB5PR05MB1254.eurprd05.prod.outlook.com; FPR:; SPF:None; PTR:InfoNoRecords; MX:1; A:1; LANG:en; received-spf: None (protection.outlook.com: mellanox.com does not designate permitted sender hosts) spamdiagnosticoutput: 1:99 spamdiagnosticmetadata: NSPM Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-OriginatorOrg: Mellanox.com X-MS-Exchange-CrossTenant-Network-Message-Id: af739c1e-b659-4109-50a7-08d51bf154e4 X-MS-Exchange-CrossTenant-originalarrivaltime: 25 Oct 2017 21:42:46.7016 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: a652971c-7d2e-4d9b-a6a4-d149256f461b X-MS-Exchange-Transport-CrossTenantHeadersStamped: DB5PR05MB1254 Subject: Re: [dpdk-dev] [PATCH v2 2/7] net/mlx4: inline more Tx functions X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 25 Oct 2017 21:42:49 -0000 Hi Adrien, On Wednesday, October 25, 2017 7:50 PM, Adrien Mazarguil wrote: >=20 > Hi Ophir, >=20 > On Mon, Oct 23, 2017 at 02:21:55PM +0000, Ophir Munk wrote: > > Change functions to inline on Tx fast path to improve performance > > > > Inside the inline function call other functions to handle "unlikely" > > cases such that the inline function code footprint is small. > > > > Signed-off-by: Ophir Munk >=20 > Reading this, it's like adding __rte_always_inline improves performance a= t > all, which I doubt unless you can show proof through performance results. >=20 > When in doubt, leave it to the compiler, the static keyword is usually en= ough > of a hint. Too much forced inlining may actually be harmful. >=20 > What this patch really does is splitting the heavy lookup/registration fu= nction > in two halves with one small static inline function for the lookup part t= hat > calls the separate registration part in the unlikely event MR is not alre= ady > registered. >=20 > Thankfully the compiler doesn't inline the large registration function ba= ck, > which results in the perceived performance improvement for the time being= , > however there is no guarantee it won't happen in the future (you didn't u= se > the noinline keyword on the registration function for that). >=20 > Therefore I have a bunch of comments and suggestions, see below. >=20 > > --- > > drivers/net/mlx4/mlx4_rxtx.c | 43 > > ++++++------------------------------ > > drivers/net/mlx4/mlx4_rxtx.h | 52 > > +++++++++++++++++++++++++++++++++++++++++++- > > 2 files changed, 58 insertions(+), 37 deletions(-) > > > > diff --git a/drivers/net/mlx4/mlx4_rxtx.c > > b/drivers/net/mlx4/mlx4_rxtx.c index 011ea79..ae37f9b 100644 > > --- a/drivers/net/mlx4/mlx4_rxtx.c > > +++ b/drivers/net/mlx4/mlx4_rxtx.c > > @@ -220,54 +220,25 @@ mlx4_txq_complete(struct txq *txq) > > return 0; > > } > > > > -/** > > - * Get memory pool (MP) from mbuf. If mbuf is indirect, the pool from > > which > > - * the cloned mbuf is allocated is returned instead. > > - * > > - * @param buf > > - * Pointer to mbuf. > > - * > > - * @return > > - * Memory pool where data is located for given mbuf. > > - */ > > -static struct rte_mempool * > > -mlx4_txq_mb2mp(struct rte_mbuf *buf) > > -{ > > - if (unlikely(RTE_MBUF_INDIRECT(buf))) > > - return rte_mbuf_from_indirect(buf)->pool; > > - return buf->pool; > > -} > > > > /** > > - * Get memory region (MR) <-> memory pool (MP) association from txq- > >mp2mr[]. > > - * Add MP to txq->mp2mr[] if it's not registered yet. If mp2mr[] is > > full, > > - * remove an entry first. > > + * Add memory region (MR) <-> memory pool (MP) association to txq- > >mp2mr[]. > > + * If mp2mr[] is full, remove an entry first. > > * > > * @param txq > > * Pointer to Tx queue structure. > > * @param[in] mp > > - * Memory pool for which a memory region lkey must be returned. > > + * Memory pool for which a memory region lkey must be added > > + * @param[in] i > > + * Index in memory pool (MP) where to add memory region (MR) > > * > > * @return > > - * mr->lkey on success, (uint32_t)-1 on failure. > > + * Added mr->lkey on success, (uint32_t)-1 on failure. > > */ > > -uint32_t > > -mlx4_txq_mp2mr(struct txq *txq, struct rte_mempool *mp) > > +uint32_t mlx4_txq_add_mr(struct txq *txq, struct rte_mempool *mp, > > +uint32_t i) > > { > > - unsigned int i; > > struct ibv_mr *mr; > > > > - for (i =3D 0; (i !=3D RTE_DIM(txq->mp2mr)); ++i) { > > - if (unlikely(txq->mp2mr[i].mp =3D=3D NULL)) { > > - /* Unknown MP, add a new MR for it. */ > > - break; > > - } > > - if (txq->mp2mr[i].mp =3D=3D mp) { > > - assert(txq->mp2mr[i].lkey !=3D (uint32_t)-1); > > - assert(txq->mp2mr[i].mr->lkey =3D=3D txq- > >mp2mr[i].lkey); > > - return txq->mp2mr[i].lkey; > > - } > > - } > > /* Add a new entry, register MR first. */ > > DEBUG("%p: discovered new memory pool \"%s\" (%p)", > > (void *)txq, mp->name, (void *)mp); diff --git > > a/drivers/net/mlx4/mlx4_rxtx.h b/drivers/net/mlx4/mlx4_rxtx.h index > > e10bbca..719ef45 100644 > > --- a/drivers/net/mlx4/mlx4_rxtx.h > > +++ b/drivers/net/mlx4/mlx4_rxtx.h > > @@ -53,6 +53,7 @@ > > > > #include "mlx4.h" > > #include "mlx4_prm.h" > > +#include "mlx4_utils.h" >=20 > Why? >=20 > > > > /** Rx queue counters. */ > > struct mlx4_rxq_stats { > > @@ -160,7 +161,6 @@ void mlx4_rx_queue_release(void *dpdk_rxq); > > > > /* mlx4_rxtx.c */ > > > > -uint32_t mlx4_txq_mp2mr(struct txq *txq, struct rte_mempool *mp); > > uint16_t mlx4_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, > > uint16_t pkts_n); > > uint16_t mlx4_rx_burst(void *dpdk_rxq, struct rte_mbuf **pkts, @@ > > -169,6 +169,8 @@ uint16_t mlx4_tx_burst_removed(void *dpdk_txq, > struct rte_mbuf **pkts, > > uint16_t pkts_n); > > uint16_t mlx4_rx_burst_removed(void *dpdk_rxq, struct rte_mbuf **pkts, > > uint16_t pkts_n); > > +uint32_t mlx4_txq_add_mr(struct txq *txq, struct rte_mempool *mp, > > + unsigned int i); > > > > /* mlx4_txq.c */ > > > > @@ -177,4 +179,52 @@ int mlx4_tx_queue_setup(struct rte_eth_dev > *dev, uint16_t idx, > > const struct rte_eth_txconf *conf); void > > mlx4_tx_queue_release(void *dpdk_txq); > > > > +/** > > + * Get memory pool (MP) from mbuf. If mbuf is indirect, the pool from > > +which > > + * the cloned mbuf is allocated is returned instead. > > + * > > + * @param buf > > + * Pointer to mbuf. > > + * > > + * @return > > + * Memory pool where data is located for given mbuf. > > + */ > > +static __rte_always_inline struct rte_mempool * mlx4_txq_mb2mp(struct > > +rte_mbuf *buf) { > > + if (unlikely(RTE_MBUF_INDIRECT(buf))) > > + return rte_mbuf_from_indirect(buf)->pool; > > + return buf->pool; > > +} > > + > > +/** > > + * Get memory region (MR) <-> memory pool (MP) association from txq- > >mp2mr[]. > > + * Call mlx4_txq_add_mr() if MP is not registered yet. > > + * > > + * @param txq > > + * Pointer to Tx queue structure. > > + * @param[in] mp > > + * Memory pool for which a memory region lkey must be returned. > > + * > > + * @return > > + * mr->lkey on success, (uint32_t)-1 on failure. > > + */ > > +static __rte_always_inline uint32_t >=20 > Note __rte_always_inline is defined in rte_common.h and should be > explicitly included (however don't do that, see below). >=20 > > +mlx4_txq_mp2mr(struct txq *txq, struct rte_mempool *mp) { > > + unsigned int i; > > + > > + for (i =3D 0; (i !=3D RTE_DIM(txq->mp2mr)); ++i) { > > + if (unlikely(txq->mp2mr[i].mp =3D=3D NULL)) { > > + /* Unknown MP, add a new MR for it. */ > > + break; > > + } > > + if (txq->mp2mr[i].mp =3D=3D mp) { > > + assert(txq->mp2mr[i].lkey !=3D (uint32_t)-1); > > + assert(txq->mp2mr[i].mr->lkey =3D=3D txq- > >mp2mr[i].lkey); >=20 > assert() requires assert.h (but don't include it, see subsequent suggesti= on). >=20 > > + return txq->mp2mr[i].lkey; > > + } > > + } > > + return mlx4_txq_add_mr(txq, mp, i); > > +} > > #endif /* MLX4_RXTX_H_ */ >=20 > So as described above, these functions do not need the __rte_always_inlin= e, > please remove it. They also do not need to be located in a header file; t= he > reason it's the case for their mlx5 counterparts is that they have to be = shared > between vectorized/non-vectorized code. No such requirement here, you > should move them back to their original spot. >=20 Static function mlx4_txq_mp2mr() must be in a header file because it is sha= red by 2 files: mlx4_txq.c and mlx4_rxtx.c. It is not related to vectorized/non-vectorized code in mlx5. Having said that -__rte_always_inline is required as well otherwise compila= tion fails with=20 drivers/net/mlx4/mlx4_rxtx.h:200:1: error: 'mlx4_txq_mp2mr' defined but not= used [-Werror=3Dunused-function] for files which include mlx4_rxtx.h > My suggestion for this performance improvement is to move > mlx4_txq_add_mr() to a different file, mlx4_mr.c looks like a good > candidate. This fact will ensure it's never inlined and far away from the= data > path. >=20 Function mlx4_txq_add_mr() is relatively small.=20 What do you say about preceding it with __attribute((noinline)) instead of = creating a new file? > -- > Adrien Mazarguil > 6WIND