From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Chen Jing D(Mark)" Subject: =?utf-8?q?=5BPATCH=5D_doc=3A_add_Vector_FM10K_introduc?= =?utf-8?q?tions?= Date: Wed, 23 Dec 2015 16:59:43 +0800 Message-ID: <1450861183-7568-1-git-send-email-jing.d.chen@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable To: dev@dpdk.org Return-path: Received: from mga14.intel.com (mga14.intel.com [192.55.52.115]) by dpdk.org (Postfix) with ESMTP id AB2B25936 for ; Wed, 23 Dec 2015 09:59:51 +0100 (CET) List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" From: "Chen Jing D(Mark)" Add introductions on how to enable Vector FM10K Rx/Tx functions, the preconditions and assumptions on Rx/Tx configuration parameters. The new content also lists the limitations of vector, so app/customer can do better to select best Rx/Tx functions. Signed-off-by: Chen Jing D(Mark) --- doc/guides/nics/fm10k.rst | 89 +++++++++++++++++++++++++++++++++++++++= ++++++ 1 files changed, 89 insertions(+), 0 deletions(-) diff --git a/doc/guides/nics/fm10k.rst b/doc/guides/nics/fm10k.rst index 4206b7f..54b761c 100644 --- a/doc/guides/nics/fm10k.rst +++ b/doc/guides/nics/fm10k.rst @@ -34,6 +34,95 @@ FM10K Poll Mode Driver The FM10K poll mode driver library provides support for the Intel FM1000= 0 (FM10K) family of 40GbE/100GbE adapters. =20 +Vector PMD for FM10K +-------------------- +Vector PMD uses Intel=C2=AE SIMD instructions to optimize packet I/O. +It improves load/store bandwidth efficiency of L1 data cache by using a = wider +SSE/AVX register 1 (1). +The wider register gives space to hold multiple packet buffers so as to = save +instruction number when processing bulk of packets. + +There is no change to PMD API. The RX/TX handler are the only two entrie= s for +vPMD packet I/O. They are transparently registered at runtime RX/TX exec= ution +if all condition checks pass. + +1. To date, only an SSE version of FM10K vPMD is available. + To ensure that vPMD is in the binary code, ensure that the option + CONFIG_RTE_LIBRTE_FM10K_INC_VECTOR=3Dy is in the configure file. + +Some constraints apply as pre-conditions for specific optimizations on b= ulk +packet transfers. The following sections explain RX and TX constraints i= n the +vPMD. + +RX Constraints +~~~~~~~~~~~~~~ + +Prerequisites and Pre-conditions +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ +Number of descriptor ring must be power of 2. This is the assumptions fo= r +Vector RX. With this pre-condition, ring pointer can easily scroll back = to head +after hitting tail without conditional check. Besides that, Vector RX ca= n use +it to do bit mask by ``ring_size - 1``. + +Feature not Supported by Vector RX PMD +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ +Some features are not supported when trying to increase the throughput i= n vPMD. +They are: + +* IEEE1588 + +* FDIR + +* Header split + +* RX checksum offload + +Other features are supported using optional MACRO configuration. They in= clude: + +* HW VLAN strip + +* L3/L4 packet type + +To enabled by RX_OLFLAGS (RTE_LIBRTE_FM10K_RX_OLFLAGS_ENABLE=3Dy) + +To guarantee the constraint, configuration flags in dev_conf.rxmode will= be +checked: + +* hw_vlan_extend + +* hw_ip_checksum + +* header_split + +* fdir_conf->mode + +RX Burst Size +^^^^^^^^^^^^^ + +As vPMD is focused on high throughput, which processes 4 packets at a ti= me. +So it assumes that the RX burst should be greater than 4 per burst. It r= eturns +zero if using nb_pkt < 4 in the receive handler. If nb_pkt is not multip= le of +4, a floor alignment will be applied. + +TX Constraint +~~~~~~~~~~~~~ + +Feature not Supported by TX Vector PMD +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +TX vPMD only works when txq_flags is set to FM10K_SIMPLE_TX_FLAG. +This means that it does not support TX multi-segment, VLAN offload and T= X csum +offload. The following MACROs are used for these three features: + +* ETH_TXQ_FLAGS_NOMULTSEGS + +* ETH_TXQ_FLAGS_NOVLANOFFL + +* ETH_TXQ_FLAGS_NOXSUMSCTP + +* ETH_TXQ_FLAGS_NOXSUMUDP + +* ETH_TXQ_FLAGS_NOXSUMTCP =20 Limitations ----------- --=20 1.7.7.6