[dpdk-dev] [PATCH v6 3/4] app/testpmd: move pkt prepare logic into a separate function

[Pavan Nikhilesh Bhagavatula]

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.

>-----Original Message-----
>From: Lin, Xueqin <xueqin.lin at intel.com>
>Sent: Tuesday, April 9, 2019 2:58 PM
>To: Pavan Nikhilesh Bhagavatula <pbhagavatula at marvell.com>; Yigit, Ferruh
><ferruh.yigit at intel.com>
>Cc: dev at dpdk.org; Xu, Qian Q <qian.q.xu at intel.com>; Li, WenjieX A
><wenjiex.a.li at intel.com>; Wang, FengqinX <fengqinx.wang at intel.com>; Yao,
>Lei A <lei.a.yao at intel.com>; Wang, Yinan <yinan.wang at intel.com>; Jerin Jacob
>Kollanukkaran <jerinj at marvell.com>; thomas at monjalon.net;
>arybchenko at solarflare.com; Iremonger, Bernard
><bernard.iremonger at intel.com>; alialnu at mellanox.com; Zhang, Qi Z
><qi.z.zhang at 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-bounces at dpdk.org] On Behalf Of Pavan Nikhilesh
>> Bhagavatula
>> Sent: Tuesday, April 2, 2019 5:54 PM
>> To: Jerin Jacob Kollanukkaran <jerinj at marvell.com>;
>> thomas at monjalon.net; arybchenko at solarflare.com; Yigit, Ferruh
>> <ferruh.yigit at intel.com>; Iremonger, Bernard
>> <bernard.iremonger at intel.com>; alialnu at mellanox.com
>> Cc: dev at dpdk.org; Pavan Nikhilesh Bhagavatula
>> <pbhagavatula at marvell.com>
>> Subject: [dpdk-dev] [PATCH v6 3/4] app/testpmd: move pkt prepare logic
>> into a separate function
>>
>> From: Pavan Nikhilesh <pbhagavatula at marvell.com>
>>
>> Move the packet prepare logic into a separate function so that it can
>> be reused later.
>>
>> Signed-off-by: Pavan Nikhilesh <pbhagavatula at 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, &eth_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(&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;
>> -		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