From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by dpdk.org (Postfix) with ESMTP id 710D36A87 for ; Wed, 11 Mar 2015 15:13:36 +0100 (CET) Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by orsmga102.jf.intel.com with ESMTP; 11 Mar 2015 07:11:48 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.11,382,1422950400"; d="scan'208";a="678542276" Received: from irsmsx153.ger.corp.intel.com ([163.33.192.75]) by fmsmga001.fm.intel.com with ESMTP; 11 Mar 2015 07:13:35 -0700 Received: from irsmsx105.ger.corp.intel.com ([169.254.7.117]) by IRSMSX153.ger.corp.intel.com ([169.254.9.234]) with mapi id 14.03.0195.001; Wed, 11 Mar 2015 14:13:33 +0000 From: "Ananyev, Konstantin" To: Vlad Zolotarov , "dev@dpdk.org" Thread-Topic: [dpdk-dev] [PATCH v7 0/3]: Add LRO support to ixgbe PMD Thread-Index: AQHQW2jkZzI3PT2qr0SlLAIJQvySXZ0XU5eg Date: Wed, 11 Mar 2015 14:13:32 +0000 Message-ID: <2601191342CEEE43887BDE71AB977258213F57FF@irsmsx105.ger.corp.intel.com> References: <1426015891-20450-1-git-send-email-vladz@cloudius-systems.com> In-Reply-To: <1426015891-20450-1-git-send-email-vladz@cloudius-systems.com> Accept-Language: en-IE, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [163.33.239.181] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v7 0/3]: Add LRO support to ixgbe PMD X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 11 Mar 2015 14:13:37 -0000 > -----Original Message----- > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Vlad Zolotarov > Sent: Tuesday, March 10, 2015 7:31 PM > To: dev@dpdk.org > Subject: [dpdk-dev] [PATCH v7 0/3]: Add LRO support to ixgbe PMD >=20 > This series adds the missing flow for enabling the LRO in the ethdev and > adds a support for this feature in the ixgbe PMD. There is a big hope tha= t this > initiative is going to be picked up by some Intel developer that would ad= d the LRO support > to other Intel PMDs. >=20 > The series starts with some cleanup work in the code the final patch (the= actual adding of > the LRO support) is going to touch/use/change. There are still quite a fe= w issues in the ixgbe > PMD code left but they have to be a matter of a different series and I've= left a few "TODO" > remarks in the code. >=20 > The LRO ("RSC" in Intel's context) PMD completion handling code follows t= he same design as the > corresponding Linux and FreeBSD implementation: pass the aggregation's cl= uster HEAD buffer to > the NEXTP entry of the software ring till EOP is met. >=20 > HW configuration follows the corresponding specs: this feature is support= ed only by x540 and > 82599 PF devices. >=20 > The feature has been tested with seastar TCP stack with the following con= figuration on Tx side: > - MTU: 400B > - 100 concurrent TCP connections. >=20 > The results were: > - Without LRO: total throughput: 0.12Gbps, coefficient of variance: 1.= 41% > - With LRO: total throughput: 8.21Gbps, coefficient of variance: 0.= 59% >=20 > This is an almost factor 80 improvement. >=20 > New in v7: > - Free not-yet-completed RSC aggregations in rte_eth_dev_stop() flow. > - Fixed rx_bulk_alloc_allowed and rx_vec_allowed initialization: > - Don't set them to FALSE in rte_eth_dev_stop() flow - the followin= g > rte_eth_dev_start() will need them. > - Reset them to TRUE in rte_eth_dev_configure() and not in a probe(= ) flow. > This will ensure the proper behaviour if port is re-configured. That changes probably should be part of another patch you submitted: [PATCH v2 3/3] ixgbe: Unify the rx_pkt_bulk callbackinitialization ? Konstantin > - Reset the sw_ring[].mbuf entry in a bulk allocation case. > This is needed for ixgbe_rx_queue_release_mbufs(). > - _recv_pkts_lro(): added the missing memory barrier before RDT update= in a > non-bulk allocation case. > - Don't allow RSC when device is configured in an SR-IOV mode. >=20 > New in v6: > - Fix of the typo in the "bug fixes" series that broke the compilation= caused a > minor change in this follow-up series. >=20 > New in v5: > - Split the series into "bug fixes" and "all the rest" so that the for= mer could be > integrated into a 2.0 release. > - Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of rt= e_ethdev.h. > - Removed the "TODO: Remove me" comment near RTE_ETHDEV_HAS_LRO_SUPPOR= T. >=20 > New in v4: > - Remove CONFIG_RTE_ETHDEV_LRO_SUPPORT from config/common_linuxapp. > - Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h. > - As a result of "ixgbe: check rxd number to avoid mbuf leak" (352078e= 8e) Vector Rx > had to get the same treatment as Rx Bulk Alloc (see PATCH4 for more = details). >=20 > New in v3: > - ixgbe_rx_alloc_bufs(): Always reset refcnt of the buffers to 1. Othe= rwise rte_pktmbuf_free() > won't free them. >=20 > New in v2: > - Removed rte_eth_dev_data.lro_bulk_alloc and added ixgbe_hw.rx_bulk_a= lloc_allowed > instead. > - Unified the rx_pkt_bulk callback setting (a separate new patch). > - Fixed a few styling and spelling issues. >=20 > Vlad Zolotarov (3): > ixgbe: Cleanups > ixgbe: Code refactoring > ixgbe: Add LRO support >=20 > lib/librte_ether/rte_ethdev.h | 9 +- > lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 29 +- > lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 + > lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 738 ++++++++++++++++++++++++++++++= ++---- > lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 + > 5 files changed, 710 insertions(+), 77 deletions(-) >=20 > -- > 2.1.0