> -----Original Message-----
> From: Pavan Nikhilesh Bhagavatula [mailto:pbhagavat...@marvell.com]
> Sent: Tuesday, April 9, 2019 5:33 PM
> To: Lin, Xueqin <xueqin....@intel.com>; Yigit, Ferruh <ferruh.yi...@intel.com>
> Cc: dev@dpdk.org; Xu, Qian Q <qian.q...@intel.com>; Li, WenjieX A
> <wenjiex.a...@intel.com>; Wang, FengqinX <fengqinx.w...@intel.com>;
> Yao, Lei A <lei.a....@intel.com>; Wang, Yinan <yinan.w...@intel.com>;
> Jerin Jacob Kollanukkaran <jer...@marvell.com>; tho...@monjalon.net;
> arybche...@solarflare.com; Iremonger, Bernard
> <bernard.iremon...@intel.com>; alia...@mellanox.com; Zhang, Qi Z
> <qi.z.zh...@intel.com>
> Subject: RE: [dpdk-dev] [PATCH v6 3/4] app/testpmd: move pkt prepare logic
> into a separate function
>
> Hi Lin,
>
> Can you check if the following patch fixes the issue?
> http://patches.dpdk.org/patch/52395/
>
> I wasn't able to catch this earlier.
>
> Regards,
> Pavan
Hi, Pavan
With this patch, testpmd can generate packets with correct src
mac address at my side now. Thanks
BRs
Lei
>
> >-----Original Message-----
> >From: Lin, Xueqin <xueqin....@intel.com>
> >Sent: Tuesday, April 9, 2019 2:58 PM
> >To: Pavan Nikhilesh Bhagavatula <pbhagavat...@marvell.com>; Yigit,
> Ferruh
> ><ferruh.yi...@intel.com>
> >Cc: dev@dpdk.org; Xu, Qian Q <qian.q...@intel.com>; Li, WenjieX A
> ><wenjiex.a...@intel.com>; Wang, FengqinX <fengqinx.w...@intel.com>;
> Yao,
> >Lei A <lei.a....@intel.com>; Wang, Yinan <yinan.w...@intel.com>; Jerin
> Jacob
> >Kollanukkaran <jer...@marvell.com>; tho...@monjalon.net;
> >arybche...@solarflare.com; Iremonger, Bernard
> ><bernard.iremon...@intel.com>; alia...@mellanox.com; Zhang, Qi Z
> ><qi.z.zh...@intel.com>
> >Subject: [EXT] RE: [dpdk-dev] [PATCH v6 3/4] app/testpmd: move pkt
> prepare
> >logic into a separate function
> >
> >External Email
> >
> >----------------------------------------------------------------------
> >Hi NIkhilesh,
> >
> >This patchset impacts some of 19.05 rc1 txonly/burst tests on Intel NIC. If
> set
> >txonly fwd, IXIA or tester peer can't receive packets that sent from app
> >generated.
> >This is high issue, block some cases test. Detailed information as below,
> need
> >you to check it soon.
> >
> >*DPDK version: 19.05.0-rc1
> >*NIC hardware: Fortville_eagle/Fortville_spirit/Niantic
> >Environment: one NIC port connect with another NIC port, or one NIC port
> >connect with IXIA
> >
> >Test Setup
> >1. Bind port to igb_uio or vfio
> >2. On DUT, setup testpmd:
> > ./x86_64-native-linuxapp-gcc/app/testpmd -c 0x1e -n 4 -- -i --rxq=4 --
> txq=4 --
> >port-topology=loop 3. Set txonly forward, start testpmd
> >testpmd>set fwd txonly
> >testpmd>start
> >4. Dump packets from tester NIC port or IXIA, find no packets were
> received on
> >the PORT0.
> >tcpdump -i <tester_interface> -v
> >
> >Best regards,
> >Xueqin
> >
> >> -----Original Message-----
> >> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pavan Nikhilesh
> >> Bhagavatula
> >> Sent: Tuesday, April 2, 2019 5:54 PM
> >> To: Jerin Jacob Kollanukkaran <jer...@marvell.com>;
> >> tho...@monjalon.net; arybche...@solarflare.com; Yigit, Ferruh
> >> <ferruh.yi...@intel.com>; Iremonger, Bernard
> >> <bernard.iremon...@intel.com>; alia...@mellanox.com
> >> Cc: dev@dpdk.org; Pavan Nikhilesh Bhagavatula
> >> <pbhagavat...@marvell.com>
> >> Subject: [dpdk-dev] [PATCH v6 3/4] app/testpmd: move pkt prepare logic
> >> into a separate function
> >>
> >> From: Pavan Nikhilesh <pbhagavat...@marvell.com>
> >>
> >> Move the packet prepare logic into a separate function so that it can
> >> be reused later.
> >>
> >> Signed-off-by: Pavan Nikhilesh <pbhagavat...@marvell.com>
> >> ---
> >> app/test-pmd/txonly.c | 163
> >> +++++++++++++++++++++---------------------
> >> 1 file changed, 83 insertions(+), 80 deletions(-)
> >>
> >> diff --git a/app/test-pmd/txonly.c b/app/test-pmd/txonly.c index
> >> 65171c1d1..56ca0ad24 100644
> >> --- a/app/test-pmd/txonly.c
> >> +++ b/app/test-pmd/txonly.c
> >> @@ -148,6 +148,80 @@ setup_pkt_udp_ip_headers(struct ipv4_hdr
> *ip_hdr,
> >> ip_hdr->hdr_checksum = (uint16_t) ip_cksum; }
> >>
> >> +static inline bool
> >> +pkt_burst_prepare(struct rte_mbuf *pkt, struct rte_mempool *mbp,
> >> + struct ether_hdr *eth_hdr, const uint16_t vlan_tci,
> >> + const uint16_t vlan_tci_outer, const uint64_t ol_flags) {
> >> + struct rte_mbuf *pkt_segs[RTE_MAX_SEGS_PER_PKT];
> >> + uint8_t ip_var = RTE_PER_LCORE(_ip_var);
> >> + struct rte_mbuf *pkt_seg;
> >> + uint32_t nb_segs, pkt_len;
> >> + uint8_t i;
> >> +
> >> + if (unlikely(tx_pkt_split == TX_PKT_SPLIT_RND))
> >> + nb_segs = random() % tx_pkt_nb_segs + 1;
> >> + else
> >> + nb_segs = tx_pkt_nb_segs;
> >> +
> >> + if (nb_segs > 1) {
> >> + if (rte_mempool_get_bulk(mbp, (void **)pkt_segs,
> nb_segs))
> >> + return false;
> >> + }
> >> +
> >> + rte_pktmbuf_reset_headroom(pkt);
> >> + pkt->data_len = tx_pkt_seg_lengths[0];
> >> + pkt->ol_flags = ol_flags;
> >> + pkt->vlan_tci = vlan_tci;
> >> + pkt->vlan_tci_outer = vlan_tci_outer;
> >> + pkt->l2_len = sizeof(struct ether_hdr);
> >> + pkt->l3_len = sizeof(struct ipv4_hdr);
> >> +
> >> + pkt_len = pkt->data_len;
> >> + pkt_seg = pkt;
> >> + for (i = 1; i < nb_segs; i++) {
> >> + pkt_seg->next = pkt_segs[i - 1];
> >> + pkt_seg = pkt_seg->next;
> >> + pkt_seg->data_len = tx_pkt_seg_lengths[i];
> >> + pkt_len += pkt_seg->data_len;
> >> + }
> >> + pkt_seg->next = NULL; /* Last segment of packet. */
> >> + /*
> >> + * Copy headers in first packet segment(s).
> >> + */
> >> + copy_buf_to_pkt(eth_hdr, sizeof(eth_hdr), pkt, 0);
> >> + copy_buf_to_pkt(&pkt_ip_hdr, sizeof(pkt_ip_hdr), pkt,
> >> + sizeof(struct ether_hdr));
> >> + if (txonly_multi_flow) {
> >> + struct ipv4_hdr *ip_hdr;
> >> + uint32_t addr;
> >> +
> >> + ip_hdr = rte_pktmbuf_mtod_offset(pkt,
> >> + struct ipv4_hdr *,
> >> + sizeof(struct ether_hdr));
> >> + /*
> >> + * Generate multiple flows by varying IP src addr. This
> >> + * enables packets are well distributed by RSS in
> >> + * receiver side if any and txonly mode can be a decent
> >> + * packet generator for developer's quick performance
> >> + * regression test.
> >> + */
> >> + addr = (IP_DST_ADDR | (ip_var++ << 8)) + rte_lcore_id();
> >> + ip_hdr->src_addr = rte_cpu_to_be_32(addr);
> >> + }
> >> + copy_buf_to_pkt(&pkt_udp_hdr, sizeof(pkt_udp_hdr), pkt,
> >> + sizeof(struct ether_hdr) +
> >> + sizeof(struct ipv4_hdr));
> >> + /*
> >> + * Complete first mbuf of packet and append it to the
> >> + * burst of packets to be transmitted.
> >> + */
> >> + pkt->nb_segs = nb_segs;
> >> + pkt->pkt_len = pkt_len;
> >> +
> >> + return true;
> >> +}
> >> +
> >> /*
> >> * Transmit a burst of multi-segments packets.
> >> */
> >> @@ -155,10 +229,8 @@ static void
> >> pkt_burst_transmit(struct fwd_stream *fs) {
> >> struct rte_mbuf *pkts_burst[MAX_PKT_BURST];
> >> - struct rte_mbuf *pkt_segs[RTE_MAX_SEGS_PER_PKT];
> >> struct rte_port *txp;
> >> struct rte_mbuf *pkt;
> >> - struct rte_mbuf *pkt_seg;
> >> struct rte_mempool *mbp;
> >> struct ether_hdr eth_hdr;
> >> uint16_t nb_tx;
> >> @@ -166,15 +238,12 @@ pkt_burst_transmit(struct fwd_stream *fs)
> >> uint16_t vlan_tci, vlan_tci_outer;
> >> uint32_t retry;
> >> uint64_t ol_flags = 0;
> >> - uint8_t ip_var = RTE_PER_LCORE(_ip_var);
> >> - uint8_t i;
> >> uint64_t tx_offloads;
> >> #ifdef RTE_TEST_PMD_RECORD_CORE_CYCLES
> >> uint64_t start_tsc;
> >> uint64_t end_tsc;
> >> uint64_t core_cycles;
> >> #endif
> >> - uint32_t nb_segs, pkt_len;
> >>
> >> #ifdef RTE_TEST_PMD_RECORD_CORE_CYCLES
> >> start_tsc = rte_rdtsc();
> >> @@ -201,85 +270,19 @@ pkt_burst_transmit(struct fwd_stream *fs)
> >>
> >> for (nb_pkt = 0; nb_pkt < nb_pkt_per_burst; nb_pkt++) {
> >> pkt = rte_mbuf_raw_alloc(mbp);
> >> - if (pkt == NULL) {
> >> - nomore_mbuf:
> >> - if (nb_pkt == 0)
> >> - return;
> >> + if (pkt == NULL)
> >> + break;
> >> + if (unlikely(!pkt_burst_prepare(pkt, mbp, ð_hdr, vlan_tci,
> >> + vlan_tci_outer, ol_flags))) {
> >> + rte_pktmbuf_free(pkt);
> >> break;
> >> }
> >> -
> >> - /*
> >> - * Using raw alloc is good to improve performance,
> >> - * but some consumers may use the headroom and so
> >> - * decrement data_off. We need to make sure it is
> >> - * reset to default value.
> >> - */
> >> - rte_pktmbuf_reset_headroom(pkt);
> >> - pkt->data_len = tx_pkt_seg_lengths[0];
> >> - pkt_seg = pkt;
> >> -
> >> - if (tx_pkt_split == TX_PKT_SPLIT_RND)
> >> - nb_segs = random() % tx_pkt_nb_segs + 1;
> >> - else
> >> - nb_segs = tx_pkt_nb_segs;
> >> -
> >> - if (nb_segs > 1) {
> >> - if (rte_mempool_get_bulk(mbp, (void **)pkt_segs,
> >> - nb_segs)) {
> >> - rte_pktmbuf_free(pkt);
> >> - goto nomore_mbuf;
> >> - }
> >> - }
> >> -
> >> - pkt_len = pkt->data_len;
> >> - for (i = 1; i < nb_segs; i++) {
> >> - pkt_seg->next = pkt_segs[i - 1];
> >> - pkt_seg = pkt_seg->next;
> >> - pkt_seg->data_len = tx_pkt_seg_lengths[i];
> >> - pkt_len += pkt_seg->data_len;
> >> - }
> >> - pkt_seg->next = NULL; /* Last segment of packet. */
> >> -
> >> - /*
> >> - * Copy headers in first packet segment(s).
> >> - */
> >> - copy_buf_to_pkt(ð_hdr, sizeof(eth_hdr), pkt, 0);
> >> - copy_buf_to_pkt(&pkt_ip_hdr, sizeof(pkt_ip_hdr), pkt,
> >> - sizeof(struct ether_hdr));
> >> - if (txonly_multi_flow) {
> >> - struct ipv4_hdr *ip_hdr;
> >> - uint32_t addr;
> >> -
> >> - ip_hdr = rte_pktmbuf_mtod_offset(pkt,
> >> - struct ipv4_hdr *,
> >> - sizeof(struct ether_hdr));
> >> - /*
> >> - * Generate multiple flows by varying IP src addr. This
> >> - * enables packets are well distributed by RSS in
> >> - * receiver side if any and txonly mode can be a
> >> decent
> >> - * packet generator for developer's quick
> >> performance
> >> - * regression test.
> >> - */
> >> - addr = (IP_DST_ADDR | (ip_var++ << 8)) +
> >> rte_lcore_id();
> >> - ip_hdr->src_addr = rte_cpu_to_be_32(addr);
> >> - }
> >> - copy_buf_to_pkt(&pkt_udp_hdr, sizeof(pkt_udp_hdr), pkt,
> >> - sizeof(struct ether_hdr) +
> >> - sizeof(struct ipv4_hdr));
> >> -
> >> - /*
> >> - * Complete first mbuf of packet and append it to the
> >> - * burst of packets to be transmitted.
> >> - */
> >> - pkt->nb_segs = nb_segs;
> >> - pkt->pkt_len = pkt_len;
> >> - pkt->ol_flags = ol_flags;
> >> - pkt->vlan_tci = vlan_tci;
> >> - pkt->vlan_tci_outer = vlan_tci_outer;
> >> - pkt->l2_len = sizeof(struct ether_hdr);
> >> - pkt->l3_len = sizeof(struct ipv4_hdr);
> >> pkts_burst[nb_pkt] = pkt;
> >> }
> >> +
> >> + if (nb_pkt == 0)
> >> + return;
> >> +
> >> nb_tx = rte_eth_tx_burst(fs->tx_port, fs->tx_queue, pkts_burst,
> >> nb_pkt);
> >> /*
> >> * Retry if necessary
> >> --
> >> 2.21.0