From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by dpdk.org (Postfix) with ESMTP id 4657D1F7 for ; Mon, 29 Sep 2014 15:30:10 +0200 (CEST) Received: from azsmga001.ch.intel.com ([10.2.17.19]) by fmsmga101.fm.intel.com with ESMTP; 29 Sep 2014 06:36:37 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.04,620,1406617200"; d="scan'208";a="480737309" Received: from irsmsx102.ger.corp.intel.com ([163.33.3.155]) by azsmga001.ch.intel.com with ESMTP; 29 Sep 2014 06:36:28 -0700 Received: from irsmsx108.ger.corp.intel.com ([169.254.11.21]) by IRSMSX102.ger.corp.intel.com ([169.254.2.200]) with mapi id 14.03.0195.001; Mon, 29 Sep 2014 14:35:22 +0100 From: "De Lara Guarch, Pablo" To: "Ananyev, Konstantin" , "Pattan, Reshma" , "dev@dpdk.org" Thread-Topic: [dpdk-dev] [PATCH v2] distributor_app: new sample app Thread-Index: AQHP2AJi1zXXgpMcD0+yx1ATRe0wtpwThu2Q///7zgCABIFTgIAAB6iAgAAVjlA= Date: Mon, 29 Sep 2014 13:35:21 +0000 Message-ID: References: <1410869607-16842-1-git-send-email-reshma.pattan@intel.com> <1411568210-2555-1-git-send-email-reshma.pattan@intel.com> <2601191342CEEE43887BDE71AB9772582137D8B1@IRSMSX104.ger.corp.intel.com> <3AEA2BF9852C6F48A459DA490692831FE20598@IRSMSX109.ger.corp.intel.com> <2601191342CEEE43887BDE71AB97725821387654@IRSMSX104.ger.corp.intel.com> In-Reply-To: <2601191342CEEE43887BDE71AB97725821387654@IRSMSX104.ger.corp.intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [163.33.239.180] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v2] distributor_app: new sample app 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: Mon, 29 Sep 2014 13:30:11 -0000 > -----Original Message----- > From: Ananyev, Konstantin > Sent: Monday, September 29, 2014 2:07 PM > To: Pattan, Reshma; De Lara Guarch, Pablo; dev@dpdk.org > Subject: RE: [dpdk-dev] [PATCH v2] distributor_app: new sample app >=20 >=20 >=20 > > -----Original Message----- > > From: Pattan, Reshma > > Sent: Monday, September 29, 2014 1:40 PM > > To: Ananyev, Konstantin; De Lara Guarch, Pablo; dev@dpdk.org > > Subject: RE: [dpdk-dev] [PATCH v2] distributor_app: new sample app > > > > > > > > -----Original Message----- > > From: Ananyev, Konstantin > > Sent: Friday, September 26, 2014 4:52 PM > > To: De Lara Guarch, Pablo; Pattan, Reshma; dev@dpdk.org > > Subject: RE: [dpdk-dev] [PATCH v2] distributor_app: new sample app > > > > > > > > > -----Original Message----- > > > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of De Lara Guarch, > > > Pablo > > > Sent: Friday, September 26, 2014 4:12 PM > > > To: Pattan, Reshma; dev@dpdk.org > > > Subject: Re: [dpdk-dev] [PATCH v2] distributor_app: new sample app > > > > > > Hi, > > > > > > > -----Original Message----- > > > > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of reshmapa > > > > Sent: Wednesday, September 24, 2014 3:17 PM > > > > To: dev@dpdk.org > > > > Subject: [dpdk-dev] [PATCH v2] distributor_app: new sample app > > > > > > > > From: Reshma Pattan > > > > > > > > A new sample app that shows the usage of the distributor library. > > > > This app works as follows: > > > > > > > > * An RX thread runs which pulls packets from each ethernet port in = turn > > > > and passes those packets to worker using a distributor component. > > > > * The workers take the packets in turn, and determine the output po= rt > > > > for those packets using basic l2forwarding doing an xor on the so= urce > > > > port id. > > > > * The RX thread takes the returned packets from the workers and > enqueue > > > > those packets into an rte_ring structure. > > > > * A TX thread pulls the packets off the rte_ring structure and then > > > > sends each packet out the output port specified previously by the > > > > worker > > > > * Command-line option support provided only for portmask. > > > > > > > > Signed-off-by: Bruce Richardson > > > > Signed-off-by: Reshma Pattan > > > > --- > > > > examples/Makefile | 1 + > > > > examples/distributor_app/Makefile | 57 ++++ > > > > examples/distributor_app/main.c | 585 > > > > ++++++++++++++++++++++++++++++++++++++ > > > > examples/distributor_app/main.h | 46 +++ > > > > 4 files changed, 689 insertions(+) > > > > create mode 100644 examples/distributor_app/Makefile create mode > > > > 100644 examples/distributor_app/main.c create mode 100644 > > > > examples/distributor_app/main.h > > > > > > > > diff --git a/examples/Makefile b/examples/Makefile index > > > > 6245f83..2ba82b0 100644 > > > > --- a/examples/Makefile > > > > +++ b/examples/Makefile > > > > @@ -66,5 +66,6 @@ DIRS-y +=3D vhost > > > > DIRS-$(CONFIG_RTE_LIBRTE_XEN_DOM0) +=3D vhost_xen DIRS-y +=3D > vmdq > > > > DIRS-y +=3D vmdq_dcb > > > > +DIRS-$(CONFIG_RTE_LIBRTE_DISTRIBUTOR) +=3D distributor_app > > > > > > > > include $(RTE_SDK)/mk/rte.extsubdir.mk diff --git > > > > a/examples/distributor_app/Makefile > > > > b/examples/distributor_app/Makefile > > > > new file mode 100644 > > > > index 0000000..394785d > > > > --- /dev/null > > > > +++ b/examples/distributor_app/Makefile > > > > @@ -0,0 +1,57 @@ > > > > +# BSD LICENSE > > > > +# > > > > +# Copyright(c) 2010-2014 Intel Corporation. All rights reserved. > > > > +# All rights reserved. > > > > +# > > > > +# Redistribution and use in source and binary forms, with or wit= hout > > > > +# modification, are permitted provided that the following condit= ions > > > > +# are met: > > > > +# > > > > +# * Redistributions of source code must retain the above copyr= ight > > > > +# notice, this list of conditions and the following disclaim= er. > > > > +# * Redistributions in binary form must reproduce the above > copyright > > > > +# notice, this list of conditions and the following disclaim= er in > > > > +# the documentation and/or other materials provided with the > > > > +# distribution. > > > > +# * Neither the name of Intel Corporation nor the names of its > > > > +# contributors may be used to endorse or promote products > derived > > > > +# from this software without specific prior written permissi= on. > > > > +# > > > > +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND > > > > CONTRIBUTORS > > > > +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, > BUT > > > > NOT > > > > +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND > > > > FITNESS FOR > > > > +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE > > > > COPYRIGHT > > > > +# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, > > > > INCIDENTAL, > > > > +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, > BUT > > > > NOT > > > > +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; > LOSS > > > > OF USE, > > > > +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER > CAUSED > > > > AND ON ANY > > > > +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR > > > > TORT > > > > +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY > OUT OF > > > > THE USE > > > > +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH > > > > DAMAGE. > > > > + > > > > +ifeq ($(RTE_SDK),) > > > > +$(error "Please define RTE_SDK environment variable") endif > > > > + > > > > +# Default target, can be overriden by command line or environment > > > > +RTE_TARGET ?=3D x86_64-default-linuxapp-gcc > > > > > > This target is not present anymore. Change it to x86_64-native-linuxa= pp- > gcc. > > > > > > > + > > > > +include $(RTE_SDK)/mk/rte.vars.mk > > > > + > > > > +# binary name > > > > +APP =3D distributor_app > > > > + > > > > +# all source are stored in SRCS-y > > > > +SRCS-y :=3D main.c > > > > + > > > > +CFLAGS +=3D $(WERROR_FLAGS) > > > > + > > > > +# workaround for a gcc bug with noreturn attribute # > > > > +http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D12603 > > > > +ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y) CFLAGS_main.o +=3D > > > > +-Wno-return-type endif > > > > + > > > > +EXTRA_CFLAGS +=3D -O3 -Wfatal-errors > > > > + > > > > +include $(RTE_SDK)/mk/rte.extapp.mk > > > > diff --git a/examples/distributor_app/main.c > > > > b/examples/distributor_app/main.c new file mode 100644 index > > > > 0000000..628810a > > > > --- /dev/null > > > > +++ b/examples/distributor_app/main.c > > > > @@ -0,0 +1,585 @@ > > > > +/*- > > > > + * BSD LICENSE > > > > + * > > > > + * Copyright(c) 2010-2014 Intel Corporation. All rights reserved= . > > > > + * All rights reserved. > > > > + * > > > > + * Redistribution and use in source and binary forms, with or wi= thout > > > > + * modification, are permitted provided that the following condi= tions > > > > + * are met: > > > > + * > > > > + * * Redistributions of source code must retain the above copy= right > > > > + * notice, this list of conditions and the following disclai= mer. > > > > + * * Redistributions in binary form must reproduce the above > copyright > > > > + * notice, this list of conditions and the following disclai= mer in > > > > + * the documentation and/or other materials provided with th= e > > > > + * distribution. > > > > + * * Neither the name of Intel Corporation nor the names of it= s > > > > + * contributors may be used to endorse or promote products > derived > > > > + * from this software without specific prior written permiss= ion. > > > > + * > > > > + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND > > > > CONTRIBUTORS > > > > + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, > BUT > > > > NOT > > > > + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND > > > > FITNESS FOR > > > > + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE > > > > COPYRIGHT > > > > + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, > > > > INCIDENTAL, > > > > + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES > (INCLUDING, BUT > > > > NOT > > > > + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; > LOSS > > > > OF USE, > > > > + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER > CAUSED > > > > AND ON ANY > > > > + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, > OR > > > > TORT > > > > + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY > OUT OF > > > > THE USE > > > > + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH > > > > DAMAGE. > > > > + */ > > > > + > > > > +#include > > > > +#include > > > > +#include > > > > +#include > > > > +#include > > > > + > > > > +#include > > > > +#include > > > > +#include > > > > +#include > > > > +#include > > > > +#include > > > > + > > > > +#include "main.h" > > > > + > > > > +#define RX_RING_SIZE 256 > > > > +#define RX_FREE_THRESH 32 > > > > +#define RX_PTHRESH 8 > > > > +#define RX_HTHRESH 8 > > > > +#define RX_WTHRESH 0 > > > > + > > > > +#define TX_RING_SIZE 512 > > > > +#define TX_FREE_THRESH 32 > > > > +#define TX_PTHRESH 32 > > > > +#define TX_HTHRESH 0 > > > > +#define TX_WTHRESH 0 > > > > +#define TX_RSBIT_THRESH 32 > > > > +#define TX_Q_FLAGS (ETH_TXQ_FLAGS_NOMULTSEGS | > > > > ETH_TXQ_FLAGS_NOVLANOFFL |\ > > > > + ETH_TXQ_FLAGS_NOXSUMSCTP | ETH_TXQ_FLAGS_NOXSUMUDP | > > > > \ > > > > + ETH_TXQ_FLAGS_NOXSUMTCP) > > > > + > > > > +#define NUM_MBUFS ((64*1024)-1) > > > > +#define MBUF_SIZE (2048 + sizeof(struct rte_mbuf) + > > > > RTE_PKTMBUF_HEADROOM) > > > > +#define MBUF_CACHE_SIZE 250 > > > > +#define BURST_SIZE 32 > > > > +#define RTE_RING_SZ 1024 > > > > + > > > > +/* mask of enabled ports */ > > > > +static uint32_t enabled_port_mask =3D 0; > > > > + > > > > +static volatile struct app_stats { > > > > + struct { > > > > + uint64_t rx_pkts; > > > > + uint64_t returned_pkts; > > > > + uint64_t enqueued_pkts; > > > > + } rx __rte_cache_aligned; > > > > + > > > > + struct { > > > > + uint64_t dequeue_pkts; > > > > + uint64_t tx_pkts; > > > > + } tx __rte_cache_aligned; > > > > +} app_stats; > > > > + > > > > +static const struct rte_eth_conf port_conf_default =3D { > > > > + .rxmode =3D { > > > > + .mq_mode =3D ETH_MQ_RX_RSS, > > > > + .max_rx_pkt_len =3D ETHER_MAX_LEN, > > > > + .split_hdr_size =3D 0, > > > > + .header_split =3D 0, /**< Header Split disabled */ > > > > + .hw_ip_checksum =3D 0, /**< IP checksum offload enabled */ > > > > + .hw_vlan_filter =3D 0, /**< VLAN filtering disabled */ > > > > + .jumbo_frame =3D 0, /**< Jumbo Frame Support disabled */ > > > > + .hw_strip_crc =3D 0, /**< CRC stripped by hardware */ > > > > + }, > > > > + .txmode =3D { > > > > + .mq_mode =3D ETH_MQ_TX_NONE, > > > > + }, > > > > + .lpbk_mode =3D 0, > > > > + .rx_adv_conf =3D { > > > > + .rss_conf =3D { > > > > + .rss_hf =3D ETH_RSS_IPV4 | ETH_RSS_IPV6 | > > > > + ETH_RSS_IPV4_TCP | > > > > ETH_RSS_IPV4_UDP | > > > > + ETH_RSS_IPV6_TCP | > > > > ETH_RSS_IPV6_UDP, > > > > + } > > > > + }, > > > > +}; > > > > + > > > > +static const struct rte_eth_rxconf rx_conf_default =3D { > > > > + .rx_thresh =3D { > > > > + .pthresh =3D RX_PTHRESH, > > > > + .hthresh =3D RX_HTHRESH, > > > > + .wthresh =3D RX_WTHRESH, > > > > + }, > > > > + .rx_free_thresh =3D RX_FREE_THRESH, > > > > + .rx_drop_en =3D 0, > > > > +}; > > > > + > > > > +static const struct rte_eth_txconf tx_conf_default =3D { > > > > + .tx_thresh =3D { > > > > + .pthresh =3D TX_PTHRESH, > > > > + .hthresh =3D TX_HTHRESH, > > > > + .wthresh =3D TX_WTHRESH, > > > > + }, > > > > + .tx_free_thresh =3D TX_FREE_THRESH, > > > > + .tx_rs_thresh =3D TX_RSBIT_THRESH, > > > > + .txq_flags =3D TX_Q_FLAGS > > > > + > > > > +}; > > > > + > > > > +struct output_buffer { > > > > + unsigned count; > > > > + struct rte_mbuf *mbufs[BURST_SIZE]; }; > > > > + > > > > +/* > > > > + * Initialises a given port using global settings and with the rx > > > > +buffers > > > > + * coming from the mbuf_pool passed as parameter */ static inline > > > > +int port_init(uint8_t port, struct rte_mempool *mbuf_pool) { > > > > + struct rte_eth_conf port_conf =3D port_conf_default; > > > > + const uint16_t rxRings =3D 1, txRings =3D rte_lcore_count() - 1; > > > > + int retval; > > > > + uint16_t q; > > > > + > > > > + if (port >=3D rte_eth_dev_count()) > > > > + return -1; > > > > + > > > > + retval =3D rte_eth_dev_configure(port, rxRings, txRings, &port_co= nf); > > > > + if (retval !=3D 0) > > > > + return retval; > > > > + > > > > + for (q =3D 0; q < rxRings; q++) { > > > > + retval =3D rte_eth_rx_queue_setup(port, q, RX_RING_SIZE, > > > > + > > > > rte_eth_dev_socket_id(port), > > > > + &rx_conf_default, > > > > mbuf_pool); > > > > + if (retval < 0) > > > > + return retval; > > > > + } > > > > + > > > > + for (q =3D 0; q < txRings; q++) { > > > > + retval =3D rte_eth_tx_queue_setup(port, q, TX_RING_SIZE, > > > > + > > > > rte_eth_dev_socket_id(port), > > > > + &tx_conf_default); > > > > + if (retval < 0) > > > > + return retval; > > > > + } > > > > + > > > > + retval =3D rte_eth_dev_start(port); > > > > + if (retval < 0) > > > > + return retval; > > > > + > > > > + struct rte_eth_link link; > > > > + rte_eth_link_get_nowait(port, &link); > > > > + if (!link.link_status) { > > > > + sleep(1); > > > > + rte_eth_link_get_nowait(port, &link); > > > > + } > > > > + > > > > + if (!link.link_status) { > > > > + printf("Link down on port %"PRIu8"\n", port); > > > > + return 0; > > > > + } > > > > + > > > > + struct ether_addr addr; > > > > + rte_eth_macaddr_get(port, &addr); > > > > + printf("Port %u MAC: %02"PRIx8" %02"PRIx8" %02"PRIx8 > > > > + " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n", > > > > + (unsigned)port, > > > > + addr.addr_bytes[0], addr.addr_bytes[1], > > > > + addr.addr_bytes[2], addr.addr_bytes[3], > > > > + addr.addr_bytes[4], addr.addr_bytes[5]); > > > > + > > > > + rte_eth_promiscuous_enable(port); > > > > + > > > > + return 0; > > > > +} > > > > + > > > > +struct lcore_params { > > > > + unsigned worker_id; > > > > + struct rte_distributor *d; > > > > + struct rte_ring *r; > > > > +}; > > > > + > > > > +static __attribute__((noreturn)) void lcore_rx(struct lcore_params > > > > +*p) { > > > > + struct rte_distributor *d =3D p->d; > > > > + struct rte_ring *r =3D p->r; > > > > + const uint8_t nb_ports =3D rte_eth_dev_count(); > > > > + const int socket_id =3D rte_socket_id(); > > > > + uint8_t port; > > > > + > > > > + for (port =3D 0; port < nb_ports; port++) { > > > > + /* skip ports that are not enabled */ > > > > + if ((enabled_port_mask & (1 << port)) =3D=3D 0) > > > > + continue; > > > > + > > > > + if (rte_eth_dev_socket_id(port) > 0 && > > > > + rte_eth_dev_socket_id(port) !=3D socket_id) > > > > + printf("WARNING, port %u is on remote NUMA node > > > > to " > > > > + "RX thread.\n\tPerformance will not > > > > " > > > > + "be optimal.\n", port); > > > > + } > > > > + > > > > + printf("\nCore %u doing packet RX.\n", rte_lcore_id()); > > > > + port =3D 0; > > > > + for (;;) { > > > > + /* skip ports that are not enabled */ > > > > + if ((enabled_port_mask & (1 << port)) =3D=3D 0) { > > > > + if (++port =3D=3D nb_ports) > > > > + port =3D 0; > > > > + continue; > > > > + } > > > > + struct rte_mbuf *bufs[BURST_SIZE*2]; > > > > + const uint16_t nb_rx =3D rte_eth_rx_burst(port, 0, bufs, > > > > + BURST_SIZE); > > > > + app_stats.rx.rx_pkts +=3D nb_rx; > > > > + > > > > + rte_distributor_process(d, bufs, nb_rx); > > > > + const uint16_t nb_ret =3D rte_distributor_returned_pkts(d, > > > > + bufs, BURST_SIZE*2); > > > > + app_stats.rx.returned_pkts +=3D nb_ret; > > > > + if (unlikely(nb_ret =3D=3D 0)) > > > > + continue; > > > > + > > > > + uint16_t sent =3D rte_ring_enqueue_burst(r, (void *)bufs, > > > > nb_ret); > > > > + app_stats.rx.enqueued_pkts +=3D sent; > > > > + if (unlikely(sent < nb_ret)) { > > > > + printf("Packet loss due to full ring\n"); > > > > > > I would remove this printf. If core is too busy to process packets, t= his may > make performance worse. > > > Statistics should reflect this packet loss. > > > > >Probably use RTE_LOG(DEBUG, ...). > > >That's way you can switch it on/off at compile time. > > > > By default LOG_LEVEL is 8 in config/common_linuxapp, hence I see > RTE_LOG(DEBUG,...) always works. > > So until unless change the LOG-LEVEL to lesser value, RTE_LOG(DEBUG,..) > will be logged .. > > Please let me know if I can go ahead and remove the printf as of now un= til > the LOG_LEVEL is fixed. (or) I have to add some logic to log > > RTE_LOG(DEBUG,..) only once. >=20 > Sorry, I probably wasn't clear enough. > What I am saying - create something like that inside your app code: >=20 > #ifdef XXX_DEBUG > #define XXX_LOG(level, fmt, args...) \ > RTE_LOG(level, PMD, "%s(): " fmt "\n", __func__, ## args) > #else > #define XXX_LOG(level, fmt, args...) do { } while(0) > #endif >=20 > And replace your pritnf()s with it. I think we should make some changes in some logs in DPDK, and change the de= fault log level to be INFO or less, so we could use the DEBUG level for thi= ngs like this, but anyway, that would be another patch that I am thinking t= o submit some other day. Anyway, for this one, this option could work, but still that message may be= shown hundreds of thousands of times per second, considering that CPU is o= verloaded. So, should we include something like a flag here, so message is = shown just once? I know that this will only be used when DEBUG is enabled, but still that co= uld flood the screen, as it is in an infinite loop. >=20 > > > > > > > > > + while (sent < nb_ret) > > > > + rte_pktmbuf_free(bufs[sent++]); > > > > + } > > > > + if (++port =3D=3D nb_ports) > > > > + port =3D 0; > > > > + } > > > > +} > > > > + > > > > +static inline void > > > > +flush_one_port(struct output_buffer *outbuf, uint8_t outp) { > > > > + unsigned nb_tx =3D rte_eth_tx_burst(outp, 0, outbuf->mbufs, > > > > + outbuf->count); > > > > + app_stats.tx.tx_pkts +=3D nb_tx; > > > > + > > > > + if (unlikely(nb_tx < outbuf->count)) { > > > > + printf("Packet loss with tx_burst\n"); > > > > > > Same here. This is fine for debugging, but statistics are updated as = well, > so we do not need this printf. > > > > > > > + do { > > > > + rte_pktmbuf_free(outbuf->mbufs[nb_tx]); > > > > + } while (++nb_tx < outbuf->count); > > > > + } > > > > + outbuf->count =3D 0; > > > > +} > > > > +