net/i40e: add AVX2 Rx function
authorBruce Richardson <bruce.richardson@intel.com>
Tue, 9 Jan 2018 14:32:54 +0000 (14:32 +0000)
committerFerruh Yigit <ferruh.yigit@intel.com>
Tue, 16 Jan 2018 17:47:49 +0000 (18:47 +0100)
Add a new Rx function using AVX2 instructions for higher
performance. For now, this functionality is limited to platforms
with Intel Xeon Scalable Processor(SP). The function to be used
is selected at runtime, not just at compile-time.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
drivers/net/i40e/i40e_rxtx.c
drivers/net/i40e/i40e_rxtx.h
drivers/net/i40e/i40e_rxtx_vec_avx2.c

index 22e9eae..23256b7 100644 (file)
@@ -1685,7 +1685,9 @@ i40e_dev_supported_ptypes_get(struct rte_eth_dev *dev)
 #endif
            dev->rx_pkt_burst == i40e_recv_scattered_pkts ||
            dev->rx_pkt_burst == i40e_recv_scattered_pkts_vec ||
-           dev->rx_pkt_burst == i40e_recv_pkts_vec)
+           dev->rx_pkt_burst == i40e_recv_pkts_vec ||
+           dev->rx_pkt_burst == i40e_recv_scattered_pkts_vec_avx2 ||
+           dev->rx_pkt_burst == i40e_recv_pkts_vec_avx2)
                return ptypes;
        return NULL;
 }
@@ -2807,6 +2809,17 @@ i40e_set_rx_function(struct rte_eth_dev *dev)
                                     dev->data->port_id);
 
                        dev->rx_pkt_burst = i40e_recv_scattered_pkts_vec;
+#ifdef RTE_ARCH_X86
+                       /*
+                        * since AVX frequency can be different to base
+                        * frequency, limit use of AVX2 version to later
+                        * plaforms, not all those that could theoretically
+                        * run it.
+                        */
+                       if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F))
+                               dev->rx_pkt_burst =
+                                       i40e_recv_scattered_pkts_vec_avx2;
+#endif
                } else {
                        PMD_INIT_LOG(DEBUG, "Using a Scattered with bulk "
                                           "allocation callback (port=%d).",
@@ -2826,6 +2839,16 @@ i40e_set_rx_function(struct rte_eth_dev *dev)
                             dev->data->port_id);
 
                dev->rx_pkt_burst = i40e_recv_pkts_vec;
+#ifdef RTE_ARCH_X86
+               /*
+                * since AVX frequency can be different to base
+                * frequency, limit use of AVX2 version to later
+                * plaforms, not all those that could theoretically
+                * run it.
+                */
+               if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F))
+                       dev->rx_pkt_burst = i40e_recv_pkts_vec_avx2;
+#endif
        } else if (ad->rx_bulk_alloc_allowed) {
                PMD_INIT_LOG(DEBUG, "Rx Burst Bulk Alloc Preconditions are "
                                    "satisfied. Rx Burst Bulk Alloc function "
@@ -2846,7 +2869,9 @@ i40e_set_rx_function(struct rte_eth_dev *dev)
        if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
                rx_using_sse =
                        (dev->rx_pkt_burst == i40e_recv_scattered_pkts_vec ||
-                        dev->rx_pkt_burst == i40e_recv_pkts_vec);
+                        dev->rx_pkt_burst == i40e_recv_pkts_vec ||
+                        dev->rx_pkt_burst == i40e_recv_scattered_pkts_vec_avx2 ||
+                        dev->rx_pkt_burst == i40e_recv_pkts_vec_avx2);
 
                for (i = 0; i < dev->data->nb_rx_queues; i++) {
                        struct i40e_rx_queue *rxq = dev->data->rx_queues[i];
@@ -3019,6 +3044,22 @@ i40e_recv_scattered_pkts_vec(
        return 0;
 }
 
+uint16_t __attribute__((weak))
+i40e_recv_pkts_vec_avx2(void __rte_unused *rx_queue,
+                       struct rte_mbuf __rte_unused **rx_pkts,
+                       uint16_t __rte_unused nb_pkts)
+{
+       return 0;
+}
+
+uint16_t __attribute__((weak))
+i40e_recv_scattered_pkts_vec_avx2(void __rte_unused *rx_queue,
+                       struct rte_mbuf __rte_unused **rx_pkts,
+                       uint16_t __rte_unused nb_pkts)
+{
+       return 0;
+}
+
 int __attribute__((weak))
 i40e_rxq_vec_setup(struct i40e_rx_queue __rte_unused *rxq)
 {
index 49114cf..34cd792 100644 (file)
@@ -227,6 +227,10 @@ void i40e_set_tx_function_flag(struct rte_eth_dev *dev,
 void i40e_set_tx_function(struct rte_eth_dev *dev);
 void i40e_set_default_ptype_table(struct rte_eth_dev *dev);
 void i40e_set_default_pctype_table(struct rte_eth_dev *dev);
+uint16_t i40e_recv_pkts_vec_avx2(void *rx_queue, struct rte_mbuf **rx_pkts,
+       uint16_t nb_pkts);
+uint16_t i40e_recv_scattered_pkts_vec_avx2(void *rx_queue,
+       struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
 uint16_t i40e_xmit_pkts_vec_avx2(void *tx_queue, struct rte_mbuf **tx_pkts,
        uint16_t nb_pkts);
 
index 9d3b6cc..ea6e771 100644 (file)
 #pragma GCC diagnostic ignored "-Wcast-qual"
 #endif
 
+static inline void
+i40e_rxq_rearm(struct i40e_rx_queue *rxq)
+{
+       int i;
+       uint16_t rx_id;
+       volatile union i40e_rx_desc *rxdp;
+       struct i40e_rx_entry *rxep = &rxq->sw_ring[rxq->rxrearm_start];
+
+       rxdp = rxq->rx_ring + rxq->rxrearm_start;
+
+       /* Pull 'n' more MBUFs into the software ring */
+       if (rte_mempool_get_bulk(rxq->mp,
+                                (void *)rxep,
+                                RTE_I40E_RXQ_REARM_THRESH) < 0) {
+               if (rxq->rxrearm_nb + RTE_I40E_RXQ_REARM_THRESH >=
+                   rxq->nb_rx_desc) {
+                       __m128i dma_addr0;
+                       dma_addr0 = _mm_setzero_si128();
+                       for (i = 0; i < RTE_I40E_DESCS_PER_LOOP; i++) {
+                               rxep[i].mbuf = &rxq->fake_mbuf;
+                               _mm_store_si128((__m128i *)&rxdp[i].read,
+                                               dma_addr0);
+                       }
+               }
+               rte_eth_devices[rxq->port_id].data->rx_mbuf_alloc_failed +=
+                       RTE_I40E_RXQ_REARM_THRESH;
+               return;
+       }
+
+#ifndef RTE_LIBRTE_I40E_16BYTE_RX_DESC
+       struct rte_mbuf *mb0, *mb1;
+       __m128i dma_addr0, dma_addr1;
+       __m128i hdr_room = _mm_set_epi64x(RTE_PKTMBUF_HEADROOM,
+                       RTE_PKTMBUF_HEADROOM);
+       /* Initialize the mbufs in vector, process 2 mbufs in one loop */
+       for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH; i += 2, rxep += 2) {
+               __m128i vaddr0, vaddr1;
+
+               mb0 = rxep[0].mbuf;
+               mb1 = rxep[1].mbuf;
+
+               /* load buf_addr(lo 64bit) and buf_physaddr(hi 64bit) */
+               RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_physaddr) !=
+                               offsetof(struct rte_mbuf, buf_addr) + 8);
+               vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
+               vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
+
+               /* convert pa to dma_addr hdr/data */
+               dma_addr0 = _mm_unpackhi_epi64(vaddr0, vaddr0);
+               dma_addr1 = _mm_unpackhi_epi64(vaddr1, vaddr1);
+
+               /* add headroom to pa values */
+               dma_addr0 = _mm_add_epi64(dma_addr0, hdr_room);
+               dma_addr1 = _mm_add_epi64(dma_addr1, hdr_room);
+
+               /* flush desc with pa dma_addr */
+               _mm_store_si128((__m128i *)&rxdp++->read, dma_addr0);
+               _mm_store_si128((__m128i *)&rxdp++->read, dma_addr1);
+       }
+#else
+       struct rte_mbuf *mb0, *mb1, *mb2, *mb3;
+       __m256i dma_addr0_1, dma_addr2_3;
+       __m256i hdr_room = _mm256_set1_epi64x(RTE_PKTMBUF_HEADROOM);
+       /* Initialize the mbufs in vector, process 4 mbufs in one loop */
+       for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH;
+                       i += 4, rxep += 4, rxdp += 4) {
+               __m128i vaddr0, vaddr1, vaddr2, vaddr3;
+               __m256i vaddr0_1, vaddr2_3;
+
+               mb0 = rxep[0].mbuf;
+               mb1 = rxep[1].mbuf;
+               mb2 = rxep[2].mbuf;
+               mb3 = rxep[3].mbuf;
+
+               /* load buf_addr(lo 64bit) and buf_physaddr(hi 64bit) */
+               RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_physaddr) !=
+                               offsetof(struct rte_mbuf, buf_addr) + 8);
+               vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
+               vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
+               vaddr2 = _mm_loadu_si128((__m128i *)&mb2->buf_addr);
+               vaddr3 = _mm_loadu_si128((__m128i *)&mb3->buf_addr);
+
+               /*
+                * merge 0 & 1, by casting 0 to 256-bit and inserting 1
+                * into the high lanes. Similarly for 2 & 3
+                */
+               vaddr0_1 = _mm256_inserti128_si256(
+                               _mm256_castsi128_si256(vaddr0), vaddr1, 1);
+               vaddr2_3 = _mm256_inserti128_si256(
+                               _mm256_castsi128_si256(vaddr2), vaddr3, 1);
+
+               /* convert pa to dma_addr hdr/data */
+               dma_addr0_1 = _mm256_unpackhi_epi64(vaddr0_1, vaddr0_1);
+               dma_addr2_3 = _mm256_unpackhi_epi64(vaddr2_3, vaddr2_3);
+
+               /* add headroom to pa values */
+               dma_addr0_1 = _mm256_add_epi64(dma_addr0_1, hdr_room);
+               dma_addr2_3 = _mm256_add_epi64(dma_addr2_3, hdr_room);
+
+               /* flush desc with pa dma_addr */
+               _mm256_store_si256((__m256i *)&rxdp->read, dma_addr0_1);
+               _mm256_store_si256((__m256i *)&(rxdp + 2)->read, dma_addr2_3);
+       }
+
+#endif
+
+       rxq->rxrearm_start += RTE_I40E_RXQ_REARM_THRESH;
+       if (rxq->rxrearm_start >= rxq->nb_rx_desc)
+               rxq->rxrearm_start = 0;
+
+       rxq->rxrearm_nb -= RTE_I40E_RXQ_REARM_THRESH;
+
+       rx_id = (uint16_t)((rxq->rxrearm_start == 0) ?
+                            (rxq->nb_rx_desc - 1) : (rxq->rxrearm_start - 1));
+
+       /* Update the tail pointer on the NIC */
+       I40E_PCI_REG_WRITE(rxq->qrx_tail, rx_id);
+}
+
+#define PKTLEN_SHIFT     10
+
+static inline uint16_t
+_recv_raw_pkts_vec_avx2(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts,
+               uint16_t nb_pkts, uint8_t *split_packet)
+{
+#define RTE_I40E_DESCS_PER_LOOP_AVX 8
+
+       const uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl;
+       const __m256i mbuf_init = _mm256_set_epi64x(0, 0,
+                       0, rxq->mbuf_initializer);
+       struct i40e_rx_entry *sw_ring = &rxq->sw_ring[rxq->rx_tail];
+       volatile union i40e_rx_desc *rxdp = rxq->rx_ring + rxq->rx_tail;
+       const int avx_aligned = ((rxq->rx_tail & 1) == 0);
+       rte_prefetch0(rxdp);
+
+       /* nb_pkts has to be floor-aligned to RTE_I40E_DESCS_PER_LOOP_AVX */
+       nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_I40E_DESCS_PER_LOOP_AVX);
+
+       /* See if we need to rearm the RX queue - gives the prefetch a bit
+        * of time to act
+        */
+       while (rxq->rxrearm_nb > RTE_I40E_RXQ_REARM_THRESH)
+               i40e_rxq_rearm(rxq);
+
+       /* Before we start moving massive data around, check to see if
+        * there is actually a packet available
+        */
+       if (!(rxdp->wb.qword1.status_error_len &
+                       rte_cpu_to_le_32(1 << I40E_RX_DESC_STATUS_DD_SHIFT)))
+               return 0;
+
+       /* constants used in processing loop */
+       const __m256i crc_adjust = _mm256_set_epi16(
+                       /* first descriptor */
+                       0, 0, 0,       /* ignore non-length fields */
+                       -rxq->crc_len, /* sub crc on data_len */
+                       0,             /* ignore high-16bits of pkt_len */
+                       -rxq->crc_len, /* sub crc on pkt_len */
+                       0, 0,          /* ignore pkt_type field */
+                       /* second descriptor */
+                       0, 0, 0,       /* ignore non-length fields */
+                       -rxq->crc_len, /* sub crc on data_len */
+                       0,             /* ignore high-16bits of pkt_len */
+                       -rxq->crc_len, /* sub crc on pkt_len */
+                       0, 0           /* ignore pkt_type field */
+       );
+
+       /* 8 packets DD mask, LSB in each 32-bit value */
+       const __m256i dd_check = _mm256_set1_epi32(1);
+
+       /* 8 packets EOP mask, second-LSB in each 32-bit value */
+       const __m256i eop_check = _mm256_slli_epi32(dd_check,
+                       I40E_RX_DESC_STATUS_EOF_SHIFT);
+
+       /* mask to shuffle from desc. to mbuf (2 descriptors)*/
+       const __m256i shuf_msk = _mm256_set_epi8(
+                       /* first descriptor */
+                       7, 6, 5, 4,  /* octet 4~7, 32bits rss */
+                       3, 2,        /* octet 2~3, low 16 bits vlan_macip */
+                       15, 14,      /* octet 15~14, 16 bits data_len */
+                       0xFF, 0xFF,  /* skip high 16 bits pkt_len, zero out */
+                       15, 14,      /* octet 15~14, low 16 bits pkt_len */
+                       0xFF, 0xFF,  /* pkt_type set as unknown */
+                       0xFF, 0xFF,  /*pkt_type set as unknown */
+                       /* second descriptor */
+                       7, 6, 5, 4,  /* octet 4~7, 32bits rss */
+                       3, 2,        /* octet 2~3, low 16 bits vlan_macip */
+                       15, 14,      /* octet 15~14, 16 bits data_len */
+                       0xFF, 0xFF,  /* skip high 16 bits pkt_len, zero out */
+                       15, 14,      /* octet 15~14, low 16 bits pkt_len */
+                       0xFF, 0xFF,  /* pkt_type set as unknown */
+                       0xFF, 0xFF   /*pkt_type set as unknown */
+       );
+       /*
+        * compile-time check the above crc and shuffle layout is correct.
+        * NOTE: the first field (lowest address) is given last in set_epi
+        * calls above.
+        */
+       RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, pkt_len) !=
+                       offsetof(struct rte_mbuf, rx_descriptor_fields1) + 4);
+       RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, data_len) !=
+                       offsetof(struct rte_mbuf, rx_descriptor_fields1) + 8);
+       RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, vlan_tci) !=
+                       offsetof(struct rte_mbuf, rx_descriptor_fields1) + 10);
+       RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, hash) !=
+                       offsetof(struct rte_mbuf, rx_descriptor_fields1) + 12);
+
+       /* Status/Error flag masks */
+       /*
+        * mask everything except RSS, flow director and VLAN flags
+        * bit2 is for VLAN tag, bit11 for flow director indication
+        * bit13:12 for RSS indication. Bits 3-5 of error
+        * field (bits 22-24) are for IP/L4 checksum errors
+        */
+       const __m256i flags_mask = _mm256_set1_epi32(
+                       (1 << 2) | (1 << 11) | (3 << 12) | (7 << 22));
+       /*
+        * data to be shuffled by result of flag mask. If VLAN bit is set,
+        * (bit 2), then position 4 in this array will be used in the
+        * destination
+        */
+       const __m256i vlan_flags_shuf = _mm256_set_epi32(
+                       0, 0, PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED, 0,
+                       0, 0, PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED, 0);
+       /*
+        * data to be shuffled by result of flag mask, shifted down 11.
+        * If RSS/FDIR bits are set, shuffle moves appropriate flags in
+        * place.
+        */
+       const __m256i rss_flags_shuf = _mm256_set_epi8(
+                       0, 0, 0, 0, 0, 0, 0, 0,
+                       PKT_RX_RSS_HASH | PKT_RX_FDIR, PKT_RX_RSS_HASH, 0, 0,
+                       0, 0, PKT_RX_FDIR, 0, /* end up 128-bits */
+                       0, 0, 0, 0, 0, 0, 0, 0,
+                       PKT_RX_RSS_HASH | PKT_RX_FDIR, PKT_RX_RSS_HASH, 0, 0,
+                       0, 0, PKT_RX_FDIR, 0);
+
+       /*
+        * data to be shuffled by the result of the flags mask shifted by 22
+        * bits.  This gives use the l3_l4 flags.
+        */
+       const __m256i l3_l4_flags_shuf = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0,
+                       /* shift right 1 bit to make sure it not exceed 255 */
+                       (PKT_RX_EIP_CKSUM_BAD | PKT_RX_L4_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD) >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_EIP_CKSUM_BAD | PKT_RX_L4_CKSUM_BAD) >> 1,
+                       (PKT_RX_EIP_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD) >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_EIP_CKSUM_BAD) >> 1,
+                       (PKT_RX_L4_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD) >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_L4_CKSUM_BAD) >> 1,
+                       PKT_RX_IP_CKSUM_BAD >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_L4_CKSUM_GOOD) >> 1,
+                       /* second 128-bits */
+                       0, 0, 0, 0, 0, 0, 0, 0,
+                       (PKT_RX_EIP_CKSUM_BAD | PKT_RX_L4_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD) >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_EIP_CKSUM_BAD | PKT_RX_L4_CKSUM_BAD) >> 1,
+                       (PKT_RX_EIP_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD) >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_EIP_CKSUM_BAD) >> 1,
+                       (PKT_RX_L4_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD) >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_L4_CKSUM_BAD) >> 1,
+                       PKT_RX_IP_CKSUM_BAD >> 1,
+                       (PKT_RX_IP_CKSUM_GOOD | PKT_RX_L4_CKSUM_GOOD) >> 1);
+
+       const __m256i cksum_mask = _mm256_set1_epi32(
+                       PKT_RX_IP_CKSUM_GOOD | PKT_RX_IP_CKSUM_BAD |
+                       PKT_RX_L4_CKSUM_GOOD | PKT_RX_L4_CKSUM_BAD |
+                       PKT_RX_EIP_CKSUM_BAD);
+
+       RTE_SET_USED(avx_aligned); /* for 32B descriptors we don't use this */
+
+       uint16_t i, received;
+       for (i = 0, received = 0; i < nb_pkts;
+                       i += RTE_I40E_DESCS_PER_LOOP_AVX,
+                       rxdp += RTE_I40E_DESCS_PER_LOOP_AVX) {
+               /* step 1, copy over 8 mbuf pointers to rx_pkts array */
+               _mm256_storeu_si256((void *)&rx_pkts[i],
+                               _mm256_loadu_si256((void *)&sw_ring[i]));
+#ifdef RTE_ARCH_X86_64
+               _mm256_storeu_si256((void *)&rx_pkts[i + 4],
+                               _mm256_loadu_si256((void *)&sw_ring[i + 4]));
+#endif
+
+               __m256i raw_desc0_1, raw_desc2_3, raw_desc4_5, raw_desc6_7;
+#ifdef RTE_LIBRTE_I40E_16BYTE_RX_DESC
+               /* for AVX we need alignment otherwise loads are not atomic */
+               if (avx_aligned) {
+                       /* load in descriptors, 2 at a time, in reverse order */
+                       raw_desc6_7 = _mm256_load_si256((void *)(rxdp + 6));
+                       rte_compiler_barrier();
+                       raw_desc4_5 = _mm256_load_si256((void *)(rxdp + 4));
+                       rte_compiler_barrier();
+                       raw_desc2_3 = _mm256_load_si256((void *)(rxdp + 2));
+                       rte_compiler_barrier();
+                       raw_desc0_1 = _mm256_load_si256((void *)(rxdp + 0));
+               } else
+#endif
+               do {
+                       const __m128i raw_desc7 = _mm_load_si128((void *)(rxdp + 7));
+                       rte_compiler_barrier();
+                       const __m128i raw_desc6 = _mm_load_si128((void *)(rxdp + 6));
+                       rte_compiler_barrier();
+                       const __m128i raw_desc5 = _mm_load_si128((void *)(rxdp + 5));
+                       rte_compiler_barrier();
+                       const __m128i raw_desc4 = _mm_load_si128((void *)(rxdp + 4));
+                       rte_compiler_barrier();
+                       const __m128i raw_desc3 = _mm_load_si128((void *)(rxdp + 3));
+                       rte_compiler_barrier();
+                       const __m128i raw_desc2 = _mm_load_si128((void *)(rxdp + 2));
+                       rte_compiler_barrier();
+                       const __m128i raw_desc1 = _mm_load_si128((void *)(rxdp + 1));
+                       rte_compiler_barrier();
+                       const __m128i raw_desc0 = _mm_load_si128((void *)(rxdp + 0));
+
+                       raw_desc6_7 = _mm256_inserti128_si256(
+                                       _mm256_castsi128_si256(raw_desc6), raw_desc7, 1);
+                       raw_desc4_5 = _mm256_inserti128_si256(
+                                       _mm256_castsi128_si256(raw_desc4), raw_desc5, 1);
+                       raw_desc2_3 = _mm256_inserti128_si256(
+                                       _mm256_castsi128_si256(raw_desc2), raw_desc3, 1);
+                       raw_desc0_1 = _mm256_inserti128_si256(
+                                       _mm256_castsi128_si256(raw_desc0), raw_desc1, 1);
+               } while (0);
+
+               if (split_packet) {
+                       int j;
+                       for (j = 0; j < RTE_I40E_DESCS_PER_LOOP_AVX; j++)
+                               rte_mbuf_prefetch_part2(rx_pkts[i + j]);
+               }
+
+               /*
+                * convert descriptors 4-7 into mbufs, adjusting length and
+                * re-arranging fields. Then write into the mbuf
+                */
+               const __m256i len6_7 = _mm256_slli_epi32(raw_desc6_7, PKTLEN_SHIFT);
+               const __m256i len4_5 = _mm256_slli_epi32(raw_desc4_5, PKTLEN_SHIFT);
+               const __m256i desc6_7 = _mm256_blend_epi16(raw_desc6_7, len6_7, 0x80);
+               const __m256i desc4_5 = _mm256_blend_epi16(raw_desc4_5, len4_5, 0x80);
+               __m256i mb6_7 = _mm256_shuffle_epi8(desc6_7, shuf_msk);
+               __m256i mb4_5 = _mm256_shuffle_epi8(desc4_5, shuf_msk);
+               mb6_7 = _mm256_add_epi16(mb6_7, crc_adjust);
+               mb4_5 = _mm256_add_epi16(mb4_5, crc_adjust);
+               /*
+                * to get packet types, shift 64-bit values down 30 bits
+                * and so ptype is in lower 8-bits in each
+                */
+               const __m256i ptypes6_7 = _mm256_srli_epi64(desc6_7, 30);
+               const __m256i ptypes4_5 = _mm256_srli_epi64(desc4_5, 30);
+               const uint8_t ptype7 = _mm256_extract_epi8(ptypes6_7, 24);
+               const uint8_t ptype6 = _mm256_extract_epi8(ptypes6_7, 8);
+               const uint8_t ptype5 = _mm256_extract_epi8(ptypes4_5, 24);
+               const uint8_t ptype4 = _mm256_extract_epi8(ptypes4_5, 8);
+               mb6_7 = _mm256_insert_epi32(mb6_7, ptype_tbl[ptype7], 4);
+               mb6_7 = _mm256_insert_epi32(mb6_7, ptype_tbl[ptype6], 0);
+               mb4_5 = _mm256_insert_epi32(mb4_5, ptype_tbl[ptype5], 4);
+               mb4_5 = _mm256_insert_epi32(mb4_5, ptype_tbl[ptype4], 0);
+               /* merge the status bits into one register */
+               const __m256i status4_7 = _mm256_unpackhi_epi32(desc6_7,
+                               desc4_5);
+
+               /*
+                * convert descriptors 0-3 into mbufs, adjusting length and
+                * re-arranging fields. Then write into the mbuf
+                */
+               const __m256i len2_3 = _mm256_slli_epi32(raw_desc2_3, PKTLEN_SHIFT);
+               const __m256i len0_1 = _mm256_slli_epi32(raw_desc0_1, PKTLEN_SHIFT);
+               const __m256i desc2_3 = _mm256_blend_epi16(raw_desc2_3, len2_3, 0x80);
+               const __m256i desc0_1 = _mm256_blend_epi16(raw_desc0_1, len0_1, 0x80);
+               __m256i mb2_3 = _mm256_shuffle_epi8(desc2_3, shuf_msk);
+               __m256i mb0_1 = _mm256_shuffle_epi8(desc0_1, shuf_msk);
+               mb2_3 = _mm256_add_epi16(mb2_3, crc_adjust);
+               mb0_1 = _mm256_add_epi16(mb0_1, crc_adjust);
+               /* get the packet types */
+               const __m256i ptypes2_3 = _mm256_srli_epi64(desc2_3, 30);
+               const __m256i ptypes0_1 = _mm256_srli_epi64(desc0_1, 30);
+               const uint8_t ptype3 = _mm256_extract_epi8(ptypes2_3, 24);
+               const uint8_t ptype2 = _mm256_extract_epi8(ptypes2_3, 8);
+               const uint8_t ptype1 = _mm256_extract_epi8(ptypes0_1, 24);
+               const uint8_t ptype0 = _mm256_extract_epi8(ptypes0_1, 8);
+               mb2_3 = _mm256_insert_epi32(mb2_3, ptype_tbl[ptype3], 4);
+               mb2_3 = _mm256_insert_epi32(mb2_3, ptype_tbl[ptype2], 0);
+               mb0_1 = _mm256_insert_epi32(mb0_1, ptype_tbl[ptype1], 4);
+               mb0_1 = _mm256_insert_epi32(mb0_1, ptype_tbl[ptype0], 0);
+               /* merge the status bits into one register */
+               const __m256i status0_3 = _mm256_unpackhi_epi32(desc2_3,
+                               desc0_1);
+
+               /*
+                * take the two sets of status bits and merge to one
+                * After merge, the packets status flags are in the
+                * order (hi->lo): [1, 3, 5, 7, 0, 2, 4, 6]
+                */
+               __m256i status0_7 = _mm256_unpacklo_epi64(status4_7,
+                               status0_3);
+
+               /* now do flag manipulation */
+
+               /* get only flag/error bits we want */
+               const __m256i flag_bits = _mm256_and_si256(
+                               status0_7, flags_mask);
+               /* set vlan and rss flags */
+               const __m256i vlan_flags = _mm256_shuffle_epi8(
+                               vlan_flags_shuf, flag_bits);
+               const __m256i rss_flags = _mm256_shuffle_epi8(
+                               rss_flags_shuf, _mm256_srli_epi32(flag_bits, 11));
+               /*
+                * l3_l4_error flags, shuffle, then shift to correct adjustment
+                * of flags in flags_shuf, and finally mask out extra bits
+                */
+               __m256i l3_l4_flags = _mm256_shuffle_epi8(l3_l4_flags_shuf,
+                               _mm256_srli_epi32(flag_bits, 22));
+               l3_l4_flags = _mm256_slli_epi32(l3_l4_flags, 1);
+               l3_l4_flags = _mm256_and_si256(l3_l4_flags, cksum_mask);
+
+               /* merge flags */
+               const __m256i mbuf_flags = _mm256_or_si256(l3_l4_flags,
+                               _mm256_or_si256(rss_flags, vlan_flags));
+               /*
+                * At this point, we have the 8 sets of flags in the low 16-bits
+                * of each 32-bit value in vlan0.
+                * We want to extract these, and merge them with the mbuf init data
+                * so we can do a single write to the mbuf to set the flags
+                * and all the other initialization fields. Extracting the
+                * appropriate flags means that we have to do a shift and blend for
+                * each mbuf before we do the write. However, we can also
+                * add in the previously computed rx_descriptor fields to
+                * make a single 256-bit write per mbuf
+                */
+               /* check the structure matches expectations */
+               RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, ol_flags) !=
+                               offsetof(struct rte_mbuf, rearm_data) + 8);
+               RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, rearm_data) !=
+                               RTE_ALIGN(offsetof(struct rte_mbuf, rearm_data), 16));
+               /* build up data and do writes */
+               __m256i rearm0, rearm1, rearm2, rearm3, rearm4, rearm5,
+                               rearm6, rearm7;
+               rearm6 = _mm256_blend_epi32(mbuf_init, _mm256_slli_si256(mbuf_flags, 8), 0x04);
+               rearm4 = _mm256_blend_epi32(mbuf_init, _mm256_slli_si256(mbuf_flags, 4), 0x04);
+               rearm2 = _mm256_blend_epi32(mbuf_init, mbuf_flags, 0x04);
+               rearm0 = _mm256_blend_epi32(mbuf_init, _mm256_srli_si256(mbuf_flags, 4), 0x04);
+               /* permute to add in the rx_descriptor e.g. rss fields */
+               rearm6 = _mm256_permute2f128_si256(rearm6, mb6_7, 0x20);
+               rearm4 = _mm256_permute2f128_si256(rearm4, mb4_5, 0x20);
+               rearm2 = _mm256_permute2f128_si256(rearm2, mb2_3, 0x20);
+               rearm0 = _mm256_permute2f128_si256(rearm0, mb0_1, 0x20);
+               /* write to mbuf */
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 6]->rearm_data, rearm6);
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 4]->rearm_data, rearm4);
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 2]->rearm_data, rearm2);
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 0]->rearm_data, rearm0);
+
+               /* repeat for the odd mbufs */
+               const __m256i odd_flags = _mm256_castsi128_si256(
+                               _mm256_extracti128_si256(mbuf_flags, 1));
+               rearm7 = _mm256_blend_epi32(mbuf_init, _mm256_slli_si256(odd_flags, 8), 0x04);
+               rearm5 = _mm256_blend_epi32(mbuf_init, _mm256_slli_si256(odd_flags, 4), 0x04);
+               rearm3 = _mm256_blend_epi32(mbuf_init, odd_flags, 0x04);
+               rearm1 = _mm256_blend_epi32(mbuf_init, _mm256_srli_si256(odd_flags, 4), 0x04);
+               /* since odd mbufs are already in hi 128-bits use blend */
+               rearm7 = _mm256_blend_epi32(rearm7, mb6_7, 0xF0);
+               rearm5 = _mm256_blend_epi32(rearm5, mb4_5, 0xF0);
+               rearm3 = _mm256_blend_epi32(rearm3, mb2_3, 0xF0);
+               rearm1 = _mm256_blend_epi32(rearm1, mb0_1, 0xF0);
+               /* again write to mbufs */
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 7]->rearm_data, rearm7);
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 5]->rearm_data, rearm5);
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 3]->rearm_data, rearm3);
+               _mm256_storeu_si256((__m256i *)&rx_pkts[i + 1]->rearm_data, rearm1);
+
+               /* extract and record EOP bit */
+               if (split_packet) {
+                       const __m128i eop_mask = _mm_set1_epi16(
+                                       1 << I40E_RX_DESC_STATUS_EOF_SHIFT);
+                       const __m256i eop_bits256 = _mm256_and_si256(status0_7,
+                                       eop_check);
+                       /* pack status bits into a single 128-bit register */
+                       const __m128i eop_bits = _mm_packus_epi32(
+                                       _mm256_castsi256_si128(eop_bits256),
+                                       _mm256_extractf128_si256(eop_bits256, 1));
+                       /*
+                        * flip bits, and mask out the EOP bit, which is now
+                        * a split-packet bit i.e. !EOP, rather than EOP one.
+                        */
+                       __m128i split_bits = _mm_andnot_si128(eop_bits,
+                                       eop_mask);
+                       /*
+                        * eop bits are out of order, so we need to shuffle them
+                        * back into order again. In doing so, only use low 8
+                        * bits, which acts like another pack instruction
+                        * The original order is (hi->lo): 1,3,5,7,0,2,4,6
+                        * [Since we use epi8, the 16-bit positions are
+                        * multiplied by 2 in the eop_shuffle value.]
+                        */
+                       __m128i eop_shuffle = _mm_set_epi8(
+                                       0xFF, 0xFF, 0xFF, 0xFF, /* zero hi 64b */
+                                       0xFF, 0xFF, 0xFF, 0xFF,
+                                       8, 0, 10, 2, /* move values to lo 64b */
+                                       12, 4, 14, 6);
+                       split_bits = _mm_shuffle_epi8(split_bits, eop_shuffle);
+                       *(uint64_t *)split_packet = _mm_cvtsi128_si64(split_bits);
+                       split_packet += RTE_I40E_DESCS_PER_LOOP_AVX;
+               }
+
+               /* perform dd_check */
+               status0_7 = _mm256_and_si256(status0_7, dd_check);
+               status0_7 = _mm256_packs_epi32(status0_7,
+                               _mm256_setzero_si256());
+
+               uint64_t burst = __builtin_popcountll(_mm_cvtsi128_si64(
+                               _mm256_extracti128_si256(status0_7, 1)));
+               burst += __builtin_popcountll(_mm_cvtsi128_si64(
+                               _mm256_castsi256_si128(status0_7)));
+               received += burst;
+               if (burst != RTE_I40E_DESCS_PER_LOOP_AVX)
+                       break;
+       }
+
+       /* update tail pointers */
+       rxq->rx_tail += received;
+       rxq->rx_tail &= (rxq->nb_rx_desc - 1);
+       if ((rxq->rx_tail & 1) == 1 && received > 1) { /* keep avx2 aligned */
+               rxq->rx_tail--;
+               received--;
+       }
+       rxq->rxrearm_nb += received;
+       return received;
+}
+
+/*
+ * Notice:
+ * - nb_pkts < RTE_I40E_DESCS_PER_LOOP, just return no packet
+ */
+uint16_t
+i40e_recv_pkts_vec_avx2(void *rx_queue, struct rte_mbuf **rx_pkts,
+                  uint16_t nb_pkts)
+{
+       return _recv_raw_pkts_vec_avx2(rx_queue, rx_pkts, nb_pkts, NULL);
+}
+
+/*
+ * vPMD receive routine that reassembles single burst of 32 scattered packets
+ * Notice:
+ * - nb_pkts < RTE_I40E_DESCS_PER_LOOP, just return no packet
+ */
+static uint16_t
+i40e_recv_scattered_burst_vec_avx2(void *rx_queue, struct rte_mbuf **rx_pkts,
+                            uint16_t nb_pkts)
+{
+       struct i40e_rx_queue *rxq = rx_queue;
+       uint8_t split_flags[RTE_I40E_VPMD_RX_BURST] = {0};
+
+       /* get some new buffers */
+       uint16_t nb_bufs = _recv_raw_pkts_vec_avx2(rxq, rx_pkts, nb_pkts,
+                       split_flags);
+       if (nb_bufs == 0)
+               return 0;
+
+       /* happy day case, full burst + no packets to be joined */
+       const uint64_t *split_fl64 = (uint64_t *)split_flags;
+
+       if (rxq->pkt_first_seg == NULL &&
+                       split_fl64[0] == 0 && split_fl64[1] == 0 &&
+                       split_fl64[2] == 0 && split_fl64[3] == 0)
+               return nb_bufs;
+
+       /* reassemble any packets that need reassembly*/
+       unsigned int i = 0;
+
+       if (rxq->pkt_first_seg == NULL) {
+               /* find the first split flag, and only reassemble then*/
+               while (i < nb_bufs && !split_flags[i])
+                       i++;
+               if (i == nb_bufs)
+                       return nb_bufs;
+       }
+       return i + reassemble_packets(rxq, &rx_pkts[i], nb_bufs - i,
+               &split_flags[i]);
+}
+
+/*
+ * vPMD receive routine that reassembles scattered packets.
+ * Main receive routine that can handle arbitrary burst sizes
+ * Notice:
+ * - nb_pkts < RTE_I40E_DESCS_PER_LOOP, just return no packet
+ */
+uint16_t
+i40e_recv_scattered_pkts_vec_avx2(void *rx_queue, struct rte_mbuf **rx_pkts,
+                            uint16_t nb_pkts)
+{
+       uint16_t retval = 0;
+       while (nb_pkts > RTE_I40E_VPMD_RX_BURST) {
+               uint16_t burst = i40e_recv_scattered_burst_vec_avx2(rx_queue,
+                               rx_pkts + retval, RTE_I40E_VPMD_RX_BURST);
+               retval += burst;
+               nb_pkts -= burst;
+               if (burst < RTE_I40E_VPMD_RX_BURST)
+                       return retval;
+       }
+       return retval + i40e_recv_scattered_burst_vec_avx2(rx_queue,
+                               rx_pkts + retval, nb_pkts);
+}
+
+
 static inline void
 vtx1(volatile struct i40e_tx_desc *txdp,
                struct rte_mbuf *pkt, uint64_t flags)