X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=drivers%2Fnet%2Fenic%2Fenic_rxtx.c;h=aa33937006c2f404bfd09458e95f81c5196e4267;hb=93fb21fdbe23;hp=470d32c4af91d8f467a303ed281c28be25d844f1;hpb=097e1f1e4072c3748d2f26ce9660499f5f8b4b45;p=dpdk.git diff --git a/drivers/net/enic/enic_rxtx.c b/drivers/net/enic/enic_rxtx.c index 470d32c4af..aa33937006 100644 --- a/drivers/net/enic/enic_rxtx.c +++ b/drivers/net/enic/enic_rxtx.c @@ -1,42 +1,19 @@ -/* Copyright 2008-2016 Cisco Systems, Inc. All rights reserved. +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright 2008-2017 Cisco Systems, Inc. All rights reserved. * Copyright 2007 Nuova Systems, Inc. All rights reserved. - * - * Copyright (c) 2014, Cisco Systems, Inc. - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * - * 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 HOLDER 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 "enic_compat.h" #include "rq_enet_desc.h" #include "enic.h" +#include +#include +#include #define RTE_PMD_USE_PREFETCH @@ -129,6 +106,7 @@ enic_cq_rx_desc_n_bytes(struct cq_desc *cqd) CQ_ENET_RQ_DESC_BYTES_WRITTEN_MASK; } + static inline uint8_t enic_cq_rx_check_err(struct cq_desc *cqd) { @@ -143,112 +121,188 @@ enic_cq_rx_check_err(struct cq_desc *cqd) /* Lookup table to translate RX CQ flags to mbuf flags. */ static inline uint32_t -enic_cq_rx_flags_to_pkt_type(struct cq_desc *cqd) +enic_cq_rx_flags_to_pkt_type(struct cq_desc *cqd, uint8_t tnl) { struct cq_enet_rq_desc *cqrd = (struct cq_enet_rq_desc *)cqd; uint8_t cqrd_flags = cqrd->flags; + /* + * Odd-numbered entries are for tunnel packets. All packet type info + * applies to the inner packet, and there is no info on the outer + * packet. The outer flags in these entries exist only to avoid + * changing enic_cq_rx_to_pkt_flags(). They are cleared from mbuf + * afterwards. + * + * Also, as there is no tunnel type info (VXLAN, NVGRE, or GENEVE), set + * RTE_PTYPE_TUNNEL_GRENAT.. + */ static const uint32_t cq_type_table[128] __rte_cache_aligned = { [0x00] = RTE_PTYPE_UNKNOWN, - [0x20] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN - | RTE_PTYPE_L4_NONFRAG, - [0x22] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN - | RTE_PTYPE_L4_UDP, - [0x24] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN - | RTE_PTYPE_L4_TCP, - [0x60] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN - | RTE_PTYPE_L4_FRAG, - [0x62] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN - | RTE_PTYPE_L4_UDP, - [0x64] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN - | RTE_PTYPE_L4_TCP, - [0x10] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN - | RTE_PTYPE_L4_NONFRAG, - [0x12] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN - | RTE_PTYPE_L4_UDP, - [0x14] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN - | RTE_PTYPE_L4_TCP, - [0x50] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN - | RTE_PTYPE_L4_FRAG, - [0x52] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN - | RTE_PTYPE_L4_UDP, - [0x54] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN - | RTE_PTYPE_L4_TCP, + [0x20] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_NONFRAG, + [0x21] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_NONFRAG | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_NONFRAG, + [0x22] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_UDP, + [0x23] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_UDP | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_UDP, + [0x24] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_TCP, + [0x25] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_TCP | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_TCP, + [0x60] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_FRAG, + [0x61] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_FRAG | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_FRAG, + [0x62] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_UDP, + [0x63] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_UDP | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_UDP, + [0x64] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_TCP, + [0x65] = RTE_PTYPE_L3_IPV4_EXT_UNKNOWN | RTE_PTYPE_L4_TCP | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_TCP, + [0x10] = RTE_PTYPE_L3_IPV6_EXT_UNKNOWN | RTE_PTYPE_L4_NONFRAG, + [0x11] = RTE_PTYPE_L3_IPV6_EXT_UNKNOWN | RTE_PTYPE_L4_NONFRAG | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_NONFRAG, + [0x12] = RTE_PTYPE_L3_IPV6_EXT_UNKNOWN | RTE_PTYPE_L4_UDP, + [0x13] = RTE_PTYPE_L3_IPV6_EXT_UNKNOWN | RTE_PTYPE_L4_UDP | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_UDP, + [0x14] = RTE_PTYPE_L3_IPV6_EXT_UNKNOWN | RTE_PTYPE_L4_TCP, + [0x15] = RTE_PTYPE_L3_IPV6_EXT_UNKNOWN | RTE_PTYPE_L4_TCP | + RTE_PTYPE_TUNNEL_GRENAT | + RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN | + RTE_PTYPE_INNER_L4_TCP, /* All others reserved */ }; cqrd_flags &= CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT | CQ_ENET_RQ_DESC_FLAGS_IPV4 | CQ_ENET_RQ_DESC_FLAGS_IPV6 | CQ_ENET_RQ_DESC_FLAGS_TCP | CQ_ENET_RQ_DESC_FLAGS_UDP; - return cq_type_table[cqrd_flags]; + return cq_type_table[cqrd_flags + tnl]; } static inline void enic_cq_rx_to_pkt_flags(struct cq_desc *cqd, struct rte_mbuf *mbuf) { struct cq_enet_rq_desc *cqrd = (struct cq_enet_rq_desc *)cqd; - uint16_t ciflags, bwflags, pkt_flags = 0; - ciflags = enic_cq_rx_desc_ciflags(cqrd); + uint16_t bwflags, pkt_flags = 0, vlan_tci; bwflags = enic_cq_rx_desc_bwflags(cqrd); + vlan_tci = enic_cq_rx_desc_vlan(cqrd); - mbuf->ol_flags = 0; - - /* flags are meaningless if !EOP */ - if (unlikely(!enic_cq_rx_desc_eop(ciflags))) - goto mbuf_flags_done; - - /* VLAN stripping */ + /* VLAN STRIPPED flag. The L2 packet type updated here also */ if (bwflags & CQ_ENET_RQ_DESC_FLAGS_VLAN_STRIPPED) { - pkt_flags |= PKT_RX_VLAN_PKT | PKT_RX_VLAN_STRIPPED; - mbuf->vlan_tci = enic_cq_rx_desc_vlan(cqrd); + pkt_flags |= PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED; + mbuf->packet_type |= RTE_PTYPE_L2_ETHER; } else { - mbuf->vlan_tci = 0; + if (vlan_tci != 0) + mbuf->packet_type |= RTE_PTYPE_L2_ETHER_VLAN; + else + mbuf->packet_type |= RTE_PTYPE_L2_ETHER; } - - /* RSS flag */ - if (enic_cq_rx_desc_rss_type(cqrd)) { + mbuf->vlan_tci = vlan_tci; + + if ((cqd->type_color & CQ_DESC_TYPE_MASK) == CQ_DESC_TYPE_CLASSIFIER) { + struct cq_enet_rq_clsf_desc *clsf_cqd; + uint16_t filter_id; + clsf_cqd = (struct cq_enet_rq_clsf_desc *)cqd; + filter_id = clsf_cqd->filter_id; + if (filter_id) { + pkt_flags |= PKT_RX_FDIR; + if (filter_id != ENIC_MAGIC_FILTER_ID) { + mbuf->hash.fdir.hi = clsf_cqd->filter_id; + pkt_flags |= PKT_RX_FDIR_ID; + } + } + } else if (enic_cq_rx_desc_rss_type(cqrd)) { + /* RSS flag */ pkt_flags |= PKT_RX_RSS_HASH; mbuf->hash.rss = enic_cq_rx_desc_rss_hash(cqrd); } /* checksum flags */ - if (!enic_cq_rx_desc_csum_not_calc(cqrd) && - (mbuf->packet_type & RTE_PTYPE_L3_IPV4)) { - if (unlikely(!enic_cq_rx_desc_ipv4_csum_ok(cqrd))) - pkt_flags |= PKT_RX_IP_CKSUM_BAD; - if (mbuf->packet_type & (RTE_PTYPE_L4_UDP | RTE_PTYPE_L4_TCP)) { - if (unlikely(!enic_cq_rx_desc_tcp_udp_csum_ok(cqrd))) - pkt_flags |= PKT_RX_L4_CKSUM_BAD; + if (mbuf->packet_type & (RTE_PTYPE_L3_IPV4 | RTE_PTYPE_L3_IPV6)) { + if (!enic_cq_rx_desc_csum_not_calc(cqrd)) { + uint32_t l4_flags; + l4_flags = mbuf->packet_type & RTE_PTYPE_L4_MASK; + + /* + * When overlay offload is enabled, the NIC may + * set ipv4_csum_ok=1 if the inner packet is IPv6.. + * So, explicitly check for IPv4 before checking + * ipv4_csum_ok. + */ + if (mbuf->packet_type & RTE_PTYPE_L3_IPV4) { + if (enic_cq_rx_desc_ipv4_csum_ok(cqrd)) + pkt_flags |= PKT_RX_IP_CKSUM_GOOD; + else + pkt_flags |= PKT_RX_IP_CKSUM_BAD; + } + + if (l4_flags == RTE_PTYPE_L4_UDP || + l4_flags == RTE_PTYPE_L4_TCP) { + if (enic_cq_rx_desc_tcp_udp_csum_ok(cqrd)) + pkt_flags |= PKT_RX_L4_CKSUM_GOOD; + else + pkt_flags |= PKT_RX_L4_CKSUM_BAD; + } } } - mbuf_flags_done: mbuf->ol_flags = pkt_flags; } +/* dummy receive function to replace actual function in + * order to do safe reconfiguration operations. + */ +uint16_t +enic_dummy_recv_pkts(__rte_unused void *rx_queue, + __rte_unused struct rte_mbuf **rx_pkts, + __rte_unused uint16_t nb_pkts) +{ + return 0; +} + uint16_t enic_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) { - struct vnic_rq *rq = rx_queue; - struct enic *enic = vnic_dev_priv(rq->vdev); - unsigned int rx_id; + struct vnic_rq *sop_rq = rx_queue; + struct vnic_rq *data_rq; + struct vnic_rq *rq; + struct enic *enic = vnic_dev_priv(sop_rq->vdev); + uint16_t cq_idx; + uint16_t rq_idx; + uint16_t rq_num; struct rte_mbuf *nmb, *rxmb; - uint16_t nb_rx = 0, nb_err = 0; - uint16_t nb_hold; + uint16_t nb_rx = 0; struct vnic_cq *cq; volatile struct cq_desc *cqd_ptr; uint8_t color; + uint8_t tnl; + uint16_t seg_length; + struct rte_mbuf *first_seg = sop_rq->pkt_first_seg; + struct rte_mbuf *last_seg = sop_rq->pkt_last_seg; - cq = &enic->cq[enic_cq_rq(enic, rq->index)]; - rx_id = cq->to_clean; /* index of cqd, rqd, mbuf_table */ - cqd_ptr = (struct cq_desc *)(cq->ring.descs) + rx_id; + cq = &enic->cq[enic_cq_rq(enic, sop_rq->index)]; + cq_idx = cq->to_clean; /* index of cqd, rqd, mbuf_table */ + cqd_ptr = (struct cq_desc *)(cq->ring.descs) + cq_idx; - nb_hold = rq->rx_nb_hold; /* mbufs held by software */ + data_rq = &enic->rq[sop_rq->data_queue_idx]; while (nb_rx < nb_pkts) { volatile struct rq_enet_desc *rqd_ptr; - dma_addr_t dma_addr; struct cq_desc cqd; uint8_t packet_error; + uint16_t ciflags; /* Check for pkts available */ color = (cqd_ptr->type_color >> CQ_DESC_COLOR_SHIFT) @@ -256,9 +310,13 @@ enic_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, if (color == cq->last_color) break; - /* Get the cq descriptor and rq pointer */ + /* Get the cq descriptor and extract rq info from it */ cqd = *cqd_ptr; - rqd_ptr = (struct rq_enet_desc *)(rq->ring.descs) + rx_id; + rq_num = cqd.q_number & CQ_DESC_Q_NUM_MASK; + rq_idx = cqd.completed_index & CQ_DESC_COMP_NDX_MASK; + + rq = &enic->rq[rq_num]; + rqd_ptr = ((struct rq_enet_desc *)rq->ring.descs) + rq_idx; /* allocate a new mbuf */ nmb = rte_mbuf_raw_alloc(rq->mp); @@ -271,67 +329,114 @@ enic_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, packet_error = enic_cq_rx_check_err(&cqd); /* Get the mbuf to return and replace with one just allocated */ - rxmb = rq->mbuf_ring[rx_id]; - rq->mbuf_ring[rx_id] = nmb; + rxmb = rq->mbuf_ring[rq_idx]; + rq->mbuf_ring[rq_idx] = nmb; /* Increment cqd, rqd, mbuf_table index */ - rx_id++; - if (unlikely(rx_id == rq->ring.desc_count)) { - rx_id = 0; + cq_idx++; + if (unlikely(cq_idx == cq->ring.desc_count)) { + cq_idx = 0; cq->last_color = cq->last_color ? 0 : 1; } /* Prefetch next mbuf & desc while processing current one */ - cqd_ptr = (struct cq_desc *)(cq->ring.descs) + rx_id; + cqd_ptr = (struct cq_desc *)(cq->ring.descs) + cq_idx; rte_enic_prefetch(cqd_ptr); - rte_enic_prefetch(rq->mbuf_ring[rx_id]); - rte_enic_prefetch((struct rq_enet_desc *)(rq->ring.descs) - + rx_id); + + ciflags = enic_cq_rx_desc_ciflags( + (struct cq_enet_rq_desc *)&cqd); /* Push descriptor for newly allocated mbuf */ - dma_addr = (dma_addr_t)(nmb->buf_physaddr - + RTE_PKTMBUF_HEADROOM); - rqd_ptr->address = rte_cpu_to_le_64(dma_addr); - rqd_ptr->length_type = cpu_to_le16(nmb->buf_len - - RTE_PKTMBUF_HEADROOM); + nmb->data_off = RTE_PKTMBUF_HEADROOM; + /* + * Only the address needs to be refilled. length_type of the + * descriptor it set during initialization + * (enic_alloc_rx_queue_mbufs) and does not change. + */ + rqd_ptr->address = rte_cpu_to_le_64(nmb->buf_iova + + RTE_PKTMBUF_HEADROOM); + + /* Fill in the rest of the mbuf */ + seg_length = enic_cq_rx_desc_n_bytes(&cqd); + + if (rq->is_sop) { + first_seg = rxmb; + first_seg->pkt_len = seg_length; + } else { + first_seg->pkt_len = (uint16_t)(first_seg->pkt_len + + seg_length); + first_seg->nb_segs++; + last_seg->next = rxmb; + } + + rxmb->port = enic->port_id; + rxmb->data_len = seg_length; - /* Drop incoming bad packet */ + rq->rx_nb_hold++; + + if (!(enic_cq_rx_desc_eop(ciflags))) { + last_seg = rxmb; + continue; + } + + /* + * When overlay offload is enabled, CQ.fcoe indicates the + * packet is tunnelled. + */ + tnl = enic->overlay_offload && + (ciflags & CQ_ENET_RQ_DESC_FLAGS_FCOE) != 0; + /* cq rx flags are only valid if eop bit is set */ + first_seg->packet_type = + enic_cq_rx_flags_to_pkt_type(&cqd, tnl); + enic_cq_rx_to_pkt_flags(&cqd, first_seg); + /* Wipe the outer types set by enic_cq_rx_flags_to_pkt_type() */ + if (tnl) { + first_seg->packet_type &= ~(RTE_PTYPE_L3_MASK | + RTE_PTYPE_L4_MASK); + } if (unlikely(packet_error)) { - rte_pktmbuf_free(rxmb); - nb_err++; + rte_pktmbuf_free(first_seg); + rte_atomic64_inc(&enic->soft_stats.rx_packet_errors); continue; } - /* Fill in the rest of the mbuf */ - rxmb->data_off = RTE_PKTMBUF_HEADROOM; - rxmb->nb_segs = 1; - rxmb->next = NULL; - rxmb->port = enic->port_id; - rxmb->pkt_len = enic_cq_rx_desc_n_bytes(&cqd); - rxmb->packet_type = enic_cq_rx_flags_to_pkt_type(&cqd); - enic_cq_rx_to_pkt_flags(&cqd, rxmb); - rxmb->data_len = rxmb->pkt_len; /* prefetch mbuf data for caller */ - rte_packet_prefetch(RTE_PTR_ADD(rxmb->buf_addr, + rte_packet_prefetch(RTE_PTR_ADD(first_seg->buf_addr, RTE_PKTMBUF_HEADROOM)); /* store the mbuf address into the next entry of the array */ - rx_pkts[nb_rx++] = rxmb; + rx_pkts[nb_rx++] = first_seg; } - nb_hold += nb_rx + nb_err; - cq->to_clean = rx_id; + sop_rq->pkt_first_seg = first_seg; + sop_rq->pkt_last_seg = last_seg; + + cq->to_clean = cq_idx; + + if ((sop_rq->rx_nb_hold + data_rq->rx_nb_hold) > + sop_rq->rx_free_thresh) { + if (data_rq->in_use) { + data_rq->posted_index = + enic_ring_add(data_rq->ring.desc_count, + data_rq->posted_index, + data_rq->rx_nb_hold); + data_rq->rx_nb_hold = 0; + } + sop_rq->posted_index = enic_ring_add(sop_rq->ring.desc_count, + sop_rq->posted_index, + sop_rq->rx_nb_hold); + sop_rq->rx_nb_hold = 0; - if (nb_hold > rq->rx_free_thresh) { - rq->posted_index = enic_ring_add(rq->ring.desc_count, - rq->posted_index, nb_hold); - nb_hold = 0; rte_mb(); - iowrite32(rq->posted_index, &rq->ctrl->posted_index); + if (data_rq->in_use) + iowrite32_relaxed(data_rq->posted_index, + &data_rq->ctrl->posted_index); + rte_compiler_barrier(); + iowrite32_relaxed(sop_rq->posted_index, + &sop_rq->ctrl->posted_index); } - rq->rx_nb_hold = nb_hold; return nb_rx; } @@ -352,9 +457,16 @@ static inline void enic_free_wq_bufs(struct vnic_wq *wq, u16 completed_index) pool = ((struct rte_mbuf *)buf->mb)->pool; for (i = 0; i < nb_to_free; i++) { buf = &wq->bufs[tail_idx]; - m = (struct rte_mbuf *)(buf->mb); + m = rte_pktmbuf_prefree_seg((struct rte_mbuf *)(buf->mb)); + buf->mb = NULL; + + if (unlikely(m == NULL)) { + tail_idx = enic_ring_incr(desc_count, tail_idx); + continue; + } + if (likely(m->pool == pool)) { - ENIC_ASSERT(nb_free < ENIC_MAX_WQ_DESCS); + RTE_ASSERT(nb_free < ENIC_MAX_WQ_DESCS); free[nb_free++] = m; } else { rte_mempool_put_bulk(pool, (void *)free, nb_free); @@ -363,10 +475,10 @@ static inline void enic_free_wq_bufs(struct vnic_wq *wq, u16 completed_index) pool = m->pool; } tail_idx = enic_ring_incr(desc_count, tail_idx); - buf->mb = NULL; } - rte_mempool_put_bulk(pool, (void **)free, nb_free); + if (nb_free > 0) + rte_mempool_put_bulk(pool, (void **)free, nb_free); wq->tail_idx = tail_idx; wq->ring.desc_avail += nb_to_free; @@ -385,6 +497,39 @@ unsigned int enic_cleanup_wq(__rte_unused struct enic *enic, struct vnic_wq *wq) return 0; } +uint16_t enic_prep_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, + uint16_t nb_pkts) +{ + struct vnic_wq *wq = (struct vnic_wq *)tx_queue; + int32_t ret; + uint16_t i; + uint64_t ol_flags; + struct rte_mbuf *m; + + for (i = 0; i != nb_pkts; i++) { + m = tx_pkts[i]; + ol_flags = m->ol_flags; + if (ol_flags & wq->tx_offload_notsup_mask) { + rte_errno = -ENOTSUP; + return i; + } +#ifdef RTE_LIBRTE_ETHDEV_DEBUG + ret = rte_validate_tx_offload(m); + if (ret != 0) { + rte_errno = ret; + return i; + } +#endif + ret = rte_net_intel_cksum_prepare(m); + if (ret != 0) { + rte_errno = ret; + return i; + } + } + + return i; +} + uint16_t enic_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts) { @@ -406,44 +551,76 @@ uint16_t enic_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint8_t vlan_tag_insert; uint8_t eop; uint64_t bus_addr; + uint8_t offload_mode; + uint16_t header_len; + uint64_t tso; + rte_atomic64_t *tx_oversized; enic_cleanup_wq(enic, wq); wq_desc_avail = vnic_wq_desc_avail(wq); head_idx = wq->head_idx; desc_count = wq->ring.desc_count; ol_flags_mask = PKT_TX_VLAN_PKT | PKT_TX_IP_CKSUM | PKT_TX_L4_MASK; + tx_oversized = &enic->soft_stats.tx_oversized; nb_pkts = RTE_MIN(nb_pkts, ENIC_TX_XMIT_MAX); for (index = 0; index < nb_pkts; index++) { tx_pkt = *tx_pkts++; + pkt_len = tx_pkt->pkt_len; + data_len = tx_pkt->data_len; + ol_flags = tx_pkt->ol_flags; nb_segs = tx_pkt->nb_segs; + tso = ol_flags & PKT_TX_TCP_SEG; + + /* drop packet if it's too big to send */ + if (unlikely(!tso && pkt_len > ENIC_TX_MAX_PKT_SIZE)) { + rte_pktmbuf_free(tx_pkt); + rte_atomic64_inc(tx_oversized); + continue; + } + if (nb_segs > wq_desc_avail) { if (index > 0) goto post; goto done; } - pkt_len = tx_pkt->pkt_len; - data_len = tx_pkt->data_len; - ol_flags = tx_pkt->ol_flags; mss = 0; - vlan_id = 0; - vlan_tag_insert = 0; + vlan_id = tx_pkt->vlan_tci; + vlan_tag_insert = !!(ol_flags & PKT_TX_VLAN_PKT); bus_addr = (dma_addr_t) - (tx_pkt->buf_physaddr + tx_pkt->data_off); + (tx_pkt->buf_iova + tx_pkt->data_off); descs = (struct wq_enet_desc *)wq->ring.descs; desc_p = descs + head_idx; eop = (data_len == pkt_len); + offload_mode = WQ_ENET_OFFLOAD_MODE_CSUM; + header_len = 0; + + if (tso) { + header_len = tx_pkt->l2_len + tx_pkt->l3_len + + tx_pkt->l4_len; + + /* Drop if non-TCP packet or TSO seg size is too big */ + if (unlikely(header_len == 0 || ((tx_pkt->tso_segsz + + header_len) > ENIC_TX_MAX_PKT_SIZE))) { + rte_pktmbuf_free(tx_pkt); + rte_atomic64_inc(tx_oversized); + continue; + } - if (ol_flags & ol_flags_mask) { - if (ol_flags & PKT_TX_VLAN_PKT) { - vlan_tag_insert = 1; - vlan_id = tx_pkt->vlan_tci; + offload_mode = WQ_ENET_OFFLOAD_MODE_TSO; + mss = tx_pkt->tso_segsz; + /* For tunnel, need the size of outer+inner headers */ + if (ol_flags & PKT_TX_TUNNEL_MASK) { + header_len += tx_pkt->outer_l2_len + + tx_pkt->outer_l3_len; } + } + if ((ol_flags & ol_flags_mask) && (header_len == 0)) { if (ol_flags & PKT_TX_IP_CKSUM) mss |= ENIC_CALC_IP_CKSUM; @@ -456,8 +633,10 @@ uint16_t enic_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, } } - wq_enet_desc_enc(&desc_tmp, bus_addr, data_len, mss, 0, 0, eop, - eop, 0, vlan_tag_insert, vlan_id, 0); + + wq_enet_desc_enc(&desc_tmp, bus_addr, data_len, mss, header_len, + offload_mode, eop, eop, 0, vlan_tag_insert, + vlan_id, 0); *desc_p = desc_tmp; buf = &wq->bufs[head_idx]; @@ -473,12 +652,13 @@ uint16_t enic_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, if (tx_pkt->next == NULL) eop = 1; desc_p = descs + head_idx; - bus_addr = (dma_addr_t)(tx_pkt->buf_physaddr + bus_addr = (dma_addr_t)(tx_pkt->buf_iova + tx_pkt->data_off); wq_enet_desc_enc((struct wq_enet_desc *) &desc_tmp, bus_addr, data_len, - mss, 0, 0, eop, eop, 0, - vlan_tag_insert, vlan_id, 0); + mss, 0, offload_mode, eop, eop, + 0, vlan_tag_insert, vlan_id, + 0); *desc_p = desc_tmp; buf = &wq->bufs[head_idx]; @@ -490,7 +670,7 @@ uint16_t enic_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, } post: rte_wmb(); - iowrite32(head_idx, &wq->ctrl->posted_index); + iowrite32_relaxed(head_idx, &wq->ctrl->posted_index); done: wq->ring.desc_avail = wq_desc_avail; wq->head_idx = head_idx;