From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by dpdk.org (Postfix) with ESMTP id 0BFA27D06 for ; Thu, 14 Dec 2017 03:45:37 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 13 Dec 2017 18:45:37 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.45,399,1508828400"; d="scan'208";a="2649790" Received: from dpdk15.sh.intel.com ([10.67.111.77]) by orsmga006.jf.intel.com with ESMTP; 13 Dec 2017 18:45:35 -0800 From: Jiayu Hu To: dev@dpdk.org Cc: konstantin.ananyev@intel.com, jianfeng.tan@intel.com, junjie.j.chen@intel.com, stephen@networkplumber.org, john.mcnamara@intel.com, matvejchikov@gmail.com, Jiayu Hu Date: Thu, 14 Dec 2017 10:49:39 +0800 Message-Id: <1513219779-100115-3-git-send-email-jiayu.hu@intel.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1513219779-100115-1-git-send-email-jiayu.hu@intel.com> References: <1511579860-37020-1-git-send-email-jiayu.hu@intel.com> <1513219779-100115-1-git-send-email-jiayu.hu@intel.com> Subject: [dpdk-dev] [PATCH v2 2/2] gro: support VxLAN GRO X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Dec 2017 02:45:39 -0000 This patch adds a framework that allows GRO on tunneled packets. Furthermore, it leverages that framework to provide GRO support for VxLAN-encapsulated packets. Supported VxLAN packets must have an outer IPv4 header, and contain an inner TCP/IPv4 packet. VxLAN GRO doesn't check if input packets have correct checksums and doesn't update checksums for output packets. Additionally, it assumes the packets are complete (i.e., MF==0 && frag_off==0), when IP fragmentation is possible (i.e., DF==0). Signed-off-by: Jiayu Hu --- .../prog_guide/generic_receive_offload_lib.rst | 31 +- lib/librte_gro/Makefile | 1 + lib/librte_gro/gro_vxlan_tcp4.c | 512 +++++++++++++++++++++ lib/librte_gro/gro_vxlan_tcp4.h | 181 ++++++++ lib/librte_gro/rte_gro.c | 124 ++++- lib/librte_gro/rte_gro.h | 3 + 6 files changed, 827 insertions(+), 25 deletions(-) create mode 100644 lib/librte_gro/gro_vxlan_tcp4.c create mode 100644 lib/librte_gro/gro_vxlan_tcp4.h diff --git a/doc/guides/prog_guide/generic_receive_offload_lib.rst b/doc/guides/prog_guide/generic_receive_offload_lib.rst index f07d8f6..13280a5 100644 --- a/doc/guides/prog_guide/generic_receive_offload_lib.rst +++ b/doc/guides/prog_guide/generic_receive_offload_lib.rst @@ -57,7 +57,9 @@ assumes the packets are complete (i.e., MF==0 && frag_off==0), when IP fragmentation is possible (i.e., DF==0). Additionally, it complies RFC 6864 to process the IPv4 ID field. -Currently, the GRO library provides GRO supports for TCP/IPv4 packets. +Currently, the GRO library provides GRO supports for TCP/IPv4 packets and +VxLAN packets which contain an outer IPv4 header and an inner TCP/IPv4 +packet. Two Sets of API --------------- @@ -108,7 +110,8 @@ Reassembly Algorithm The reassembly algorithm is used for reassembling packets. In the GRO library, different GRO types can use different algorithms. In this -section, we will introduce an algorithm, which is used by TCP/IPv4 GRO. +section, we will introduce an algorithm, which is used by TCP/IPv4 GRO +and VxLAN GRO. Challenges ~~~~~~~~~~ @@ -185,6 +188,30 @@ Header fields deciding if two packets are neighbors include: - IP ID. The IP ID fields of the packets, whose DF bit is 0, should be increased by 1. +VxLAN GRO +--------- + +The table structure used by VxLAN GRO, which is in charge of processing +VxLAN packets with an outer IPv4 header and inner TCP/IPv4 packet, is +similar with that of TCP/IPv4 GRO. Differently, the header fields used +to define a VxLAN flow include: + +- outer source and destination: Ethernet and IP address, UDP port + +- VxLAN header (VNI and flag) + +- inner source and destination: Ethernet and IP address, TCP port + +Header fields deciding if packets are neighbors include: + +- Outer IP ID. The IP ID fields of the packets, whose DF bit in the outer + IPv4 header is 0, should be increased by 1. + +- Inner TCP sequence number + +- Inner IP ID. The IP ID fields of the packets, whose DF bit in the outer + IPv4 header is 0, should be increased by 1. + .. note:: We comply RFC 6864 to process the IP ID field. Specifically, we only check IP ID fields for the packets whose DF bit is 0. diff --git a/lib/librte_gro/Makefile b/lib/librte_gro/Makefile index eb423cc..0110455 100644 --- a/lib/librte_gro/Makefile +++ b/lib/librte_gro/Makefile @@ -45,6 +45,7 @@ LIBABIVER := 1 # source files SRCS-$(CONFIG_RTE_LIBRTE_GRO) += rte_gro.c SRCS-$(CONFIG_RTE_LIBRTE_GRO) += gro_tcp4.c +SRCS-$(CONFIG_RTE_LIBRTE_GRO) += gro_vxlan_tcp4.c # install this header file SYMLINK-$(CONFIG_RTE_LIBRTE_GRO)-include += rte_gro.h diff --git a/lib/librte_gro/gro_vxlan_tcp4.c b/lib/librte_gro/gro_vxlan_tcp4.c new file mode 100644 index 0000000..3269de6 --- /dev/null +++ b/lib/librte_gro/gro_vxlan_tcp4.c @@ -0,0 +1,512 @@ +/*- + * BSD LICENSE + * + * Copyright(c) 2017 Intel Corporation. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * 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. + * * 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 permission. + * + * 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 "gro_vxlan_tcp4.h" + +void * +gro_vxlan_tcp4_tbl_create(uint16_t socket_id, + uint16_t max_flow_num, + uint16_t max_item_per_flow) +{ + struct gro_vxlan_tcp4_tbl *tbl; + size_t size; + uint32_t entries_num, i; + + entries_num = max_flow_num * max_item_per_flow; + entries_num = RTE_MIN(entries_num, GRO_VXLAN_TCP4_TBL_MAX_ITEM_NUM); + + if (entries_num == 0) + return NULL; + + tbl = rte_zmalloc_socket(__func__, + sizeof(struct gro_vxlan_tcp4_tbl), + RTE_CACHE_LINE_SIZE, + socket_id); + if (tbl == NULL) + return NULL; + + size = sizeof(struct gro_vxlan_tcp4_item) * entries_num; + tbl->items = rte_zmalloc_socket(__func__, + size, + RTE_CACHE_LINE_SIZE, + socket_id); + if (tbl->items == NULL) { + rte_free(tbl); + return NULL; + } + tbl->max_item_num = entries_num; + + size = sizeof(struct gro_vxlan_tcp4_flow) * entries_num; + tbl->flows = rte_zmalloc_socket(__func__, + size, + RTE_CACHE_LINE_SIZE, + socket_id); + if (tbl->flows == NULL) { + rte_free(tbl->items); + rte_free(tbl); + return NULL; + } + + for (i = 0; i < entries_num; i++) + tbl->flows[i].start_index = INVALID_ARRAY_INDEX; + tbl->max_flow_num = entries_num; + + return tbl; +} + +void +gro_vxlan_tcp4_tbl_destroy(void *tbl) +{ + struct gro_vxlan_tcp4_tbl *vxlan_tbl = tbl; + + if (vxlan_tbl) { + rte_free(vxlan_tbl->items); + rte_free(vxlan_tbl->flows); + } + rte_free(vxlan_tbl); +} + +static inline uint32_t +find_an_empty_item(struct gro_vxlan_tcp4_tbl *tbl) +{ + uint32_t max_item_num = tbl->max_item_num, i; + + for (i = 0; i < max_item_num; i++) + if (tbl->items[i].inner_item.firstseg == NULL) + return i; + return INVALID_ARRAY_INDEX; +} + +static inline uint32_t +find_an_empty_flow(struct gro_vxlan_tcp4_tbl *tbl) +{ + uint32_t max_flow_num = tbl->max_flow_num, i; + + for (i = 0; i < max_flow_num; i++) + if (tbl->flows[i].start_index == INVALID_ARRAY_INDEX) + return i; + return INVALID_ARRAY_INDEX; +} + +static inline uint32_t +insert_new_item(struct gro_vxlan_tcp4_tbl *tbl, + struct rte_mbuf *pkt, + uint64_t start_time, + uint32_t prev_idx, + uint32_t sent_seq, + uint16_t outer_ip_id, + uint16_t ip_id, + uint8_t outer_is_atomic, + uint8_t is_atomic) +{ + uint32_t item_idx; + + item_idx = find_an_empty_item(tbl); + if (item_idx == INVALID_ARRAY_INDEX) + return INVALID_ARRAY_INDEX; + + tbl->items[item_idx].inner_item.firstseg = pkt; + tbl->items[item_idx].inner_item.lastseg = rte_pktmbuf_lastseg(pkt); + tbl->items[item_idx].inner_item.start_time = start_time; + tbl->items[item_idx].inner_item.next_pkt_idx = INVALID_ARRAY_INDEX; + tbl->items[item_idx].inner_item.sent_seq = sent_seq; + tbl->items[item_idx].inner_item.ip_id = ip_id; + tbl->items[item_idx].inner_item.nb_merged = 1; + tbl->items[item_idx].inner_item.is_atomic = is_atomic; + tbl->items[item_idx].outer_ip_id = outer_ip_id; + tbl->items[item_idx].outer_is_atomic = outer_is_atomic; + tbl->item_num++; + + /* If the previous packet exists, chain the new one with it. */ + if (prev_idx != INVALID_ARRAY_INDEX) { + tbl->items[item_idx].inner_item.next_pkt_idx = + tbl->items[prev_idx].inner_item.next_pkt_idx; + tbl->items[prev_idx].inner_item.next_pkt_idx = item_idx; + } + + return item_idx; +} + +static inline uint32_t +delete_item(struct gro_vxlan_tcp4_tbl *tbl, + uint32_t item_idx, + uint32_t prev_item_idx) +{ + uint32_t next_idx = tbl->items[item_idx].inner_item.next_pkt_idx; + + /* NULL indicates an empty item. */ + tbl->items[item_idx].inner_item.firstseg = NULL; + tbl->item_num--; + if (prev_item_idx != INVALID_ARRAY_INDEX) + tbl->items[prev_item_idx].inner_item.next_pkt_idx = next_idx; + + return next_idx; +} + +static inline uint32_t +insert_new_flow(struct gro_vxlan_tcp4_tbl *tbl, + struct vxlan_tcp4_flow_key *src, + uint32_t item_idx) +{ + struct vxlan_tcp4_flow_key *dst; + uint32_t flow_idx; + + flow_idx = find_an_empty_flow(tbl); + if (flow_idx == INVALID_ARRAY_INDEX) + return INVALID_ARRAY_INDEX; + + dst = &(tbl->flows[flow_idx].key); + + ether_addr_copy(&(src->inner_key.eth_saddr), + &(dst->inner_key.eth_saddr)); + ether_addr_copy(&(src->inner_key.eth_daddr), + &(dst->inner_key.eth_daddr)); + dst->inner_key.ip_src_addr = src->inner_key.ip_src_addr; + dst->inner_key.ip_dst_addr = src->inner_key.ip_dst_addr; + dst->inner_key.recv_ack = src->inner_key.recv_ack; + dst->inner_key.src_port = src->inner_key.src_port; + dst->inner_key.dst_port = src->inner_key.dst_port; + + dst->vxlan_hdr = src->vxlan_hdr; + ether_addr_copy(&(src->outer_eth_saddr), &(dst->outer_eth_saddr)); + ether_addr_copy(&(src->outer_eth_daddr), &(dst->outer_eth_daddr)); + dst->outer_ip_src_addr = src->outer_ip_src_addr; + dst->outer_ip_dst_addr = src->outer_ip_dst_addr; + dst->outer_src_port = src->outer_src_port; + dst->outer_dst_port = src->outer_dst_port; + + tbl->flows[flow_idx].start_index = item_idx; + tbl->flow_num++; + + return flow_idx; +} + +static inline int +is_same_vxlan_tcp4_flow(struct vxlan_tcp4_flow_key k1, + struct vxlan_tcp4_flow_key k2) +{ + return (is_same_ether_addr(&k1.outer_eth_saddr, &k2.outer_eth_saddr) && + is_same_ether_addr(&k1.outer_eth_daddr, + &k2.outer_eth_daddr) && + (k1.outer_ip_src_addr == k2.outer_ip_src_addr) && + (k1.outer_ip_dst_addr == k2.outer_ip_dst_addr) && + (k1.outer_src_port == k2.outer_src_port) && + (k1.outer_dst_port == k2.outer_dst_port) && + (k1.vxlan_hdr.vx_flags == k2.vxlan_hdr.vx_flags) && + (k1.vxlan_hdr.vx_vni == k2.vxlan_hdr.vx_vni) && + is_same_tcp4_flow(k1.inner_key, k2.inner_key)); +} + +static inline int +check_vxlan_seq_option(struct gro_vxlan_tcp4_item *item, + struct tcp_hdr *tcp_hdr, + uint32_t sent_seq, + uint16_t outer_ip_id, + uint16_t ip_id, + uint16_t tcp_hl, + uint16_t tcp_dl, + uint8_t outer_is_atomic, + uint8_t is_atomic) +{ + struct rte_mbuf *pkt = item->inner_item.firstseg; + int cmp; + uint16_t l2_offset; + + /* Don't merge packets whose outer DF bits are different. */ + if (item->outer_is_atomic ^ outer_is_atomic) + return 0; + + l2_offset = pkt->outer_l2_len + pkt->outer_l3_len; + cmp = check_seq_option(&item->inner_item, tcp_hdr, sent_seq, ip_id, + tcp_hl, tcp_dl, l2_offset, is_atomic); + if ((cmp == 1) && (outer_is_atomic || (outer_ip_id == + item->outer_ip_id + + item->inner_item.nb_merged))) + /* Append the packet. */ + return 1; + else if ((cmp == -1) && (outer_is_atomic || (outer_ip_id + 1 == + item->outer_ip_id))) + /* Prepend the packet. */ + return -1; + else + return 0; +} + +static inline int +merge_two_vxlan_tcp4_packets(struct gro_vxlan_tcp4_item *item, + struct rte_mbuf *pkt, + int cmp, + uint32_t sent_seq, + uint16_t outer_ip_id, + uint16_t ip_id) +{ + if (merge_two_tcp4_packets(&item->inner_item, pkt, cmp, sent_seq, + ip_id, pkt->outer_l2_len + + pkt->outer_l3_len)) { + item->outer_ip_id = cmp < 0 ? outer_ip_id : item->outer_ip_id; + return 1; + } else + return 0; +} + +static inline void +update_vxlan_header(struct gro_vxlan_tcp4_item *item) +{ + struct ipv4_hdr *ipv4_hdr; + struct udp_hdr *udp_hdr; + struct rte_mbuf *pkt = item->inner_item.firstseg; + uint16_t len; + + /* Update the outer IPv4 header. */ + len = pkt->pkt_len - pkt->outer_l2_len; + ipv4_hdr = (struct ipv4_hdr *)(rte_pktmbuf_mtod(pkt, char *) + + pkt->outer_l2_len); + ipv4_hdr->total_length = rte_cpu_to_be_16(len); + + /* Update the outer UDP header. */ + len -= pkt->outer_l3_len; + udp_hdr = (struct udp_hdr *)((char *)ipv4_hdr + pkt->outer_l3_len); + udp_hdr->dgram_len = rte_cpu_to_be_16(len); + + /* Update the inner IPv4 header. */ + len -= pkt->l2_len; + ipv4_hdr = (struct ipv4_hdr *)((char *)udp_hdr + pkt->l2_len); + ipv4_hdr->total_length = rte_cpu_to_be_16(len); +} + +int32_t +gro_vxlan_tcp4_reassemble(struct rte_mbuf *pkt, + struct gro_vxlan_tcp4_tbl *tbl, + uint64_t start_time) +{ + struct ether_hdr *outer_eth_hdr, *eth_hdr; + struct ipv4_hdr *outer_ipv4_hdr, *ipv4_hdr; + struct tcp_hdr *tcp_hdr; + struct udp_hdr *udp_hdr; + struct vxlan_hdr *vxlan_hdr; + uint32_t sent_seq; + uint16_t tcp_dl, frag_off, outer_ip_id, ip_id; + uint8_t outer_is_atomic, is_atomic; + + struct vxlan_tcp4_flow_key key; + uint32_t cur_idx, prev_idx, item_idx; + uint32_t i, max_flow_num; + int cmp; + uint16_t hdr_len; + + outer_eth_hdr = rte_pktmbuf_mtod(pkt, struct ether_hdr *); + outer_ipv4_hdr = (struct ipv4_hdr *)((char *)outer_eth_hdr + + pkt->outer_l2_len); + udp_hdr = (struct udp_hdr *)((char *)outer_ipv4_hdr + + pkt->outer_l3_len); + vxlan_hdr = (struct vxlan_hdr *)((char *)udp_hdr + + sizeof(struct udp_hdr)); + eth_hdr = (struct ether_hdr *)((char *)vxlan_hdr + + sizeof(struct vxlan_hdr)); + ipv4_hdr = (struct ipv4_hdr *)((char *)udp_hdr + pkt->l2_len); + tcp_hdr = (struct tcp_hdr *)((char *)ipv4_hdr + pkt->l3_len); + + /* + * Check if the inner TCP header flag sets FIN, SYN, RST, + * PSH, URG, ECE or CWR bit. + */ + if (tcp_hdr->tcp_flags != TCP_ACK_FLAG) + return -1; + hdr_len = pkt->outer_l2_len + pkt->outer_l3_len + pkt->l2_len + + pkt->l3_len + pkt->l4_len; + /* + * If the payload length is less than or equal to 0, return + * immediately. + */ + tcp_dl = pkt->pkt_len - hdr_len; + if (tcp_dl <= 0) + return -1; + + /* + * Save IP ID for the packet whose DF bit is 0. For the packet + * whose DF bit is 1, IP ID is ignored. + */ + frag_off = rte_be_to_cpu_16(outer_ipv4_hdr->fragment_offset); + outer_is_atomic = (frag_off & IPV4_HDR_DF_FLAG) == IPV4_HDR_DF_FLAG; + outer_ip_id = outer_is_atomic ? 0 : + rte_be_to_cpu_16(outer_ipv4_hdr->packet_id); + frag_off = rte_be_to_cpu_16(ipv4_hdr->fragment_offset); + is_atomic = (frag_off & IPV4_HDR_DF_FLAG) == IPV4_HDR_DF_FLAG; + ip_id = is_atomic ? 0 : rte_be_to_cpu_16(ipv4_hdr->packet_id); + + sent_seq = rte_be_to_cpu_32(tcp_hdr->sent_seq); + + ether_addr_copy(&(eth_hdr->s_addr), &(key.inner_key.eth_saddr)); + ether_addr_copy(&(eth_hdr->d_addr), &(key.inner_key.eth_daddr)); + key.inner_key.ip_src_addr = ipv4_hdr->src_addr; + key.inner_key.ip_dst_addr = ipv4_hdr->dst_addr; + key.inner_key.src_port = tcp_hdr->src_port; + key.inner_key.dst_port = tcp_hdr->dst_port; + key.inner_key.recv_ack = tcp_hdr->recv_ack; + + ether_addr_copy(&(outer_eth_hdr->s_addr), &(key.outer_eth_saddr)); + ether_addr_copy(&(outer_eth_hdr->d_addr), &(key.outer_eth_daddr)); + key.outer_ip_src_addr = outer_ipv4_hdr->src_addr; + key.outer_ip_dst_addr = outer_ipv4_hdr->dst_addr; + key.outer_src_port = udp_hdr->src_port; + key.outer_dst_port = udp_hdr->dst_port; + key.vxlan_hdr.vx_flags = vxlan_hdr->vx_flags; + key.vxlan_hdr.vx_vni = vxlan_hdr->vx_vni; + + /* Search for a matched flow. */ + max_flow_num = tbl->max_flow_num; + for (i = 0; i < max_flow_num; i++) { + if (tbl->flows[i].start_index != INVALID_ARRAY_INDEX && + is_same_vxlan_tcp4_flow(tbl->flows[i].key, + key)) + break; + } + + /* + * Can't find a matched flow. Insert a new flow and store the + * packet into the flow. + */ + if (i == tbl->max_flow_num) { + item_idx = insert_new_item(tbl, pkt, start_time, + INVALID_ARRAY_INDEX, sent_seq, outer_ip_id, + ip_id, outer_is_atomic, is_atomic); + if (item_idx == INVALID_ARRAY_INDEX) + return -1; + if (insert_new_flow(tbl, &key, item_idx) == + INVALID_ARRAY_INDEX) { + /* + * Fail to insert a new flow, so + * delete the inserted packet. + */ + delete_item(tbl, item_idx, INVALID_ARRAY_INDEX); + return -1; + } + return 0; + } + + /* Check all packets in the flow and try to find a neighbor. */ + cur_idx = tbl->flows[i].start_index; + prev_idx = cur_idx; + do { + cmp = check_vxlan_seq_option(&(tbl->items[cur_idx]), tcp_hdr, + sent_seq, outer_ip_id, ip_id, pkt->l4_len, + tcp_dl, outer_is_atomic, is_atomic); + if (cmp) { + if (merge_two_vxlan_tcp4_packets(&(tbl->items[cur_idx]), + pkt, cmp, sent_seq, + outer_ip_id, ip_id)) + return 1; + /* + * Can't merge two packets, as the packet + * length will be greater than the max value. + * Insert the packet into the flow. + */ + if (insert_new_item(tbl, pkt, start_time, prev_idx, + sent_seq, outer_ip_id, + ip_id, outer_is_atomic, + is_atomic) == + INVALID_ARRAY_INDEX) + return -1; + return 0; + } + prev_idx = cur_idx; + cur_idx = tbl->items[cur_idx].inner_item.next_pkt_idx; + } while (cur_idx != INVALID_ARRAY_INDEX); + + /* Can't find neighbor. Insert the packet into the flow. */ + if (insert_new_item(tbl, pkt, start_time, prev_idx, sent_seq, + outer_ip_id, ip_id, outer_is_atomic, + is_atomic) == INVALID_ARRAY_INDEX) + return -1; + + return 0; +} + +uint16_t +gro_vxlan_tcp4_tbl_timeout_flush(struct gro_vxlan_tcp4_tbl *tbl, + uint64_t flush_timestamp, + struct rte_mbuf **out, + uint16_t nb_out) +{ + uint16_t k = 0; + uint32_t i, j; + uint32_t max_flow_num = tbl->max_flow_num; + + for (i = 0; i < max_flow_num; i++) { + if (unlikely(tbl->flow_num == 0)) + return k; + + j = tbl->flows[i].start_index; + while (j != INVALID_ARRAY_INDEX) { + if (tbl->items[j].inner_item.start_time <= + flush_timestamp) { + out[k++] = tbl->items[j].inner_item.firstseg; + if (tbl->items[j].inner_item.nb_merged > 1) + update_vxlan_header(&(tbl->items[j])); + /* + * Delete the item and get the next packet + * index. + */ + j = delete_item(tbl, j, INVALID_ARRAY_INDEX); + tbl->flows[i].start_index = j; + if (j == INVALID_ARRAY_INDEX) + tbl->flow_num--; + + if (unlikely(k == nb_out)) + return k; + } else + /* + * The left packets in the flow won't be + * timeout. Go to check other flows. + */ + break; + } + } + return k; +} + +uint32_t +gro_vxlan_tcp4_tbl_pkt_count(void *tbl) +{ + struct gro_vxlan_tcp4_tbl *gro_tbl = tbl; + + if (gro_tbl) + return gro_tbl->item_num; + + return 0; +} diff --git a/lib/librte_gro/gro_vxlan_tcp4.h b/lib/librte_gro/gro_vxlan_tcp4.h new file mode 100644 index 0000000..6047190 --- /dev/null +++ b/lib/librte_gro/gro_vxlan_tcp4.h @@ -0,0 +1,181 @@ +/*- + * BSD LICENSE + * + * Copyright(c) 2017 Intel Corporation. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * 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. + * * 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 permission. + * + * 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. + */ + +#ifndef _GRO_VXLAN_TCP4_H_ +#define _GRO_VXLAN_TCP4_H_ + +#include "gro_tcp4.h" + +#define GRO_VXLAN_TCP4_TBL_MAX_ITEM_NUM (1024UL * 1024UL) + +/* Header fields representing a VxLAN flow */ +struct vxlan_tcp4_flow_key { + struct tcp4_flow_key inner_key; + struct vxlan_hdr vxlan_hdr; + + struct ether_addr outer_eth_saddr; + struct ether_addr outer_eth_daddr; + + uint32_t outer_ip_src_addr; + uint32_t outer_ip_dst_addr; + + /* Outer UDP ports */ + uint16_t outer_src_port; + uint16_t outer_dst_port; + +}; + +struct gro_vxlan_tcp4_flow { + struct vxlan_tcp4_flow_key key; + /* + * The index of the first item in the flow. INVALID_ARRAY_INDEX + * indicates an empty flow. + */ + uint32_t start_index; +}; + +struct gro_vxlan_tcp4_item { + struct gro_tcp4_item inner_item; + /* IP ID in the outer IPv4 header */ + uint16_t outer_ip_id; + /* Indicate if outer IPv4 ID can be ignored */ + uint8_t outer_is_atomic; +}; + +/* + * VxLAN (with an outer IPv4 header and inner TCP/IPv4 headers) + * reassembly table structure + */ +struct gro_vxlan_tcp4_tbl { + /* item array */ + struct gro_vxlan_tcp4_item *items; + /* flow array */ + struct gro_vxlan_tcp4_flow *flows; + /* current item number */ + uint32_t item_num; + /* current flow number */ + uint32_t flow_num; + /* the maximum item number */ + uint32_t max_item_num; + /* the maximum flow number */ + uint32_t max_flow_num; +}; + +/** + * This function creates a VxLAN reassembly table for VxLAN packets + * which have an outer IPv4 header and inner TCP/IPv4 headers. + * + * @param socket_id + * Socket index for allocating the reassemblt table + * @param max_flow_num + * The maximum number of flows in the table + * @param max_item_per_flow + * The maximum number of packets per flow + * + * @return + * - Return the table pointer on success. + * - Return NULL on failure. + */ +void *gro_vxlan_tcp4_tbl_create(uint16_t socket_id, + uint16_t max_flow_num, + uint16_t max_item_per_flow); + +/** + * This function destroys a VxLAN reassembly table. + * + * @param tbl + * Pointer pointing to the VxLAN reassembly table + */ +void gro_vxlan_tcp4_tbl_destroy(void *tbl); + +/** + * This function merges a VxLAN packet which has an outer IPv4 header and + * inner TCP/IPv4 headers. It doesn't process the packet, which has SYN, + * FIN, RST, PSH, CWR, ECE or URG set, or doesn't have payload. It + * returns the packet if there is no available space in the table. + * + * This function doesn't check if the packet has correct checksums and + * doesn't re-calculate checksums for the merged packet. Additionally, + * it assumes the packets are complete (i.e., MF==0 && frag_off==0), when + * IP fragmentation is possible (i.e., DF==0). + * + * @param pkt + * Packet to reassemble + * @param tbl + * VxLAN reassembly table pointer + * @start_time + * The time when the packet is inserted into the table + * + * @return + * - Return a positive value if the packet is merged. + * - Return zero if the packet isn't merged but stored in the table. + * - Return a negative value for invalid parameters. + */ +int32_t gro_vxlan_tcp4_reassemble(struct rte_mbuf *pkt, + struct gro_vxlan_tcp4_tbl *tbl, + uint64_t start_time); + +/** + * This function flushes timeout packets in the VxLAN reassembly table, + * and without updating checksums. + * + * @param tbl + * Pointer that points to a TCP GRO table. + * @param flush_timestamp + * This function flushes packets which are inserted into the table + * before or at the flush_timestamp. + * @param out + * Pointer array which is used to keep flushed packets. + * @param nb_out + * The element number in 'out'. It also determines the maximum number of + * packets that can be flushed finally. + * + * @return + * The number of flushed packets. + */ +uint16_t gro_vxlan_tcp4_tbl_timeout_flush(struct gro_vxlan_tcp4_tbl *tbl, + uint64_t flush_timestamp, + struct rte_mbuf **out, + uint16_t nb_out); + +/** + * This function returns the number of the packets in a VxLAN + * reassembly table. + * + * @param tbl + * Pointer points to a TCP/IPv4 reassembly table. + * + * @return + * The number of packets in the table. + */ +uint32_t gro_vxlan_tcp4_tbl_pkt_count(void *tbl); +#endif diff --git a/lib/librte_gro/rte_gro.c b/lib/librte_gro/rte_gro.c index d43f8e8..b96aa48 100644 --- a/lib/librte_gro/rte_gro.c +++ b/lib/librte_gro/rte_gro.c @@ -37,6 +37,7 @@ #include "rte_gro.h" #include "gro_tcp4.h" +#include "gro_vxlan_tcp4.h" typedef void *(*gro_tbl_create_fn)(uint16_t socket_id, uint16_t max_flow_num, @@ -45,15 +46,28 @@ typedef void (*gro_tbl_destroy_fn)(void *tbl); typedef uint32_t (*gro_tbl_pkt_count_fn)(void *tbl); static gro_tbl_create_fn tbl_create_fn[RTE_GRO_TYPE_MAX_NUM] = { - gro_tcp4_tbl_create, NULL}; + gro_tcp4_tbl_create, gro_vxlan_tcp4_tbl_create, NULL}; static gro_tbl_destroy_fn tbl_destroy_fn[RTE_GRO_TYPE_MAX_NUM] = { - gro_tcp4_tbl_destroy, NULL}; + gro_tcp4_tbl_destroy, gro_vxlan_tcp4_tbl_destroy, + NULL}; static gro_tbl_pkt_count_fn tbl_pkt_count_fn[RTE_GRO_TYPE_MAX_NUM] = { - gro_tcp4_tbl_pkt_count, NULL}; + gro_tcp4_tbl_pkt_count, gro_vxlan_tcp4_tbl_pkt_count, + NULL}; #define IS_IPV4_TCP_PKT(ptype) (RTE_ETH_IS_IPV4_HDR(ptype) && \ ((ptype & RTE_PTYPE_L4_TCP) == RTE_PTYPE_L4_TCP)) +#define IS_IPV4_VXLAN_TCP4_PKT(ptype) (RTE_ETH_IS_IPV4_HDR(ptype) && \ + ((ptype & RTE_PTYPE_L4_UDP) == RTE_PTYPE_L4_UDP) && \ + ((ptype & RTE_PTYPE_TUNNEL_VXLAN) == \ + RTE_PTYPE_TUNNEL_VXLAN) && \ + ((ptype & RTE_PTYPE_INNER_L4_TCP) == \ + RTE_PTYPE_INNER_L4_TCP) && \ + (((ptype & RTE_PTYPE_INNER_L3_MASK) & \ + (RTE_PTYPE_INNER_L3_IPV4 | \ + RTE_PTYPE_INNER_L3_IPV4_EXT | \ + RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN)) != 0)) + /* * GRO context structure. It keeps the table structures, which are * used to merge packets, for different GRO types. Before using @@ -139,13 +153,21 @@ rte_gro_reassemble_burst(struct rte_mbuf **pkts, struct gro_tcp4_flow tcp_flows[RTE_GRO_MAX_BURST_ITEM_NUM]; struct gro_tcp4_item tcp_items[RTE_GRO_MAX_BURST_ITEM_NUM] = {{0} }; + /* Allocate a reassembly table for VXLAN GRO. */ + struct gro_vxlan_tcp4_tbl vxlan_tbl; + struct gro_vxlan_tcp4_flow vxlan_flows[RTE_GRO_MAX_BURST_ITEM_NUM]; + struct gro_vxlan_tcp4_item vxlan_items[RTE_GRO_MAX_BURST_ITEM_NUM] = { + {{0}, 0, 0} }; + struct rte_mbuf *unprocess_pkts[nb_pkts]; uint64_t current_time; uint32_t item_num; int32_t ret; uint16_t i, unprocess_num = 0, nb_after_gro = nb_pkts; + uint8_t do_tcp4_gro = 0, do_vxlan_tcp4_gro = 0; - if ((param->gro_types & RTE_GRO_TCP_IPV4) == 0) + if ((param->gro_types & (RTE_GRO_IPV4_VXLAN_TCP_IPV4 | + RTE_GRO_TCP_IPV4)) == 0) return nb_pkts; /* Get the actual number of packets */ @@ -153,20 +175,46 @@ rte_gro_reassemble_burst(struct rte_mbuf **pkts, param->max_item_per_flow)); item_num = RTE_MIN(item_num, RTE_GRO_MAX_BURST_ITEM_NUM); - for (i = 0; i < item_num; i++) - tcp_flows[i].start_index = INVALID_ARRAY_INDEX; + if (param->gro_types & RTE_GRO_IPV4_VXLAN_TCP_IPV4) { + for (i = 0; i < item_num; i++) + vxlan_flows[i].start_index = INVALID_ARRAY_INDEX; + + vxlan_tbl.flows = vxlan_flows; + vxlan_tbl.items = vxlan_items; + vxlan_tbl.flow_num = 0; + vxlan_tbl.item_num = 0; + vxlan_tbl.max_flow_num = item_num; + vxlan_tbl.max_item_num = item_num; + do_vxlan_tcp4_gro = 1; + } - tcp_tbl.flows = tcp_flows; - tcp_tbl.items = tcp_items; - tcp_tbl.flow_num = 0; - tcp_tbl.item_num = 0; - tcp_tbl.max_flow_num = item_num; - tcp_tbl.max_item_num = item_num; + if (param->gro_types & RTE_GRO_TCP_IPV4) { + for (i = 0; i < item_num; i++) + tcp_flows[i].start_index = INVALID_ARRAY_INDEX; + + tcp_tbl.flows = tcp_flows; + tcp_tbl.items = tcp_items; + tcp_tbl.flow_num = 0; + tcp_tbl.item_num = 0; + tcp_tbl.max_flow_num = item_num; + tcp_tbl.max_item_num = item_num; + do_tcp4_gro = 1; + } current_time = rte_rdtsc(); for (i = 0; i < nb_pkts; i++) { - if (IS_IPV4_TCP_PKT(pkts[i]->packet_type)) { + if (do_vxlan_tcp4_gro && IS_IPV4_VXLAN_TCP4_PKT( + pkts[i]->packet_type)) { + ret = gro_vxlan_tcp4_reassemble(pkts[i], &vxlan_tbl, + current_time); + if (ret > 0) + /* Merge successfully */ + nb_after_gro--; + else if (ret < 0) + unprocess_pkts[unprocess_num++] = pkts[i]; + } else if (do_tcp4_gro && IS_IPV4_TCP_PKT( + pkts[i]->packet_type)) { ret = gro_tcp4_reassemble(pkts[i], &tcp_tbl, current_time); if (ret > 0) @@ -179,9 +227,16 @@ rte_gro_reassemble_burst(struct rte_mbuf **pkts, } if (nb_after_gro < nb_pkts) { + i = 0; /* Flush packets from the tables */ - i = gro_tcp4_tbl_timeout_flush(&tcp_tbl, current_time, - pkts, nb_pkts); + if (do_vxlan_tcp4_gro) { + i = gro_vxlan_tcp4_tbl_timeout_flush(&vxlan_tbl, + current_time, pkts, nb_pkts); + } + if (do_tcp4_gro) { + i += gro_tcp4_tbl_timeout_flush(&tcp_tbl, + current_time, &pkts[i], nb_pkts - i); + } /* Copy unprocessed packets */ if (unprocess_num > 0) { memcpy(&pkts[i], unprocess_pkts, @@ -200,18 +255,33 @@ rte_gro_reassemble(struct rte_mbuf **pkts, { struct rte_mbuf *unprocess_pkts[nb_pkts]; struct gro_ctx *gro_ctx = ctx; - void *tcp_tbl; + void *tcp_tbl, *vxlan_tbl; uint64_t current_time; uint16_t i, unprocess_num = 0; + uint8_t do_tcp4_gro = 0, do_vxlan_tcp4_gro = 0; - if ((gro_ctx->gro_types & RTE_GRO_TCP_IPV4) == 0) + if ((gro_ctx->gro_types & (RTE_GRO_IPV4_VXLAN_TCP_IPV4 | + RTE_GRO_TCP_IPV4)) == 0) return nb_pkts; + if (gro_ctx->gro_types & RTE_GRO_IPV4_VXLAN_TCP_IPV4) { + do_vxlan_tcp4_gro = 1; + vxlan_tbl = gro_ctx->tbls[RTE_GRO_IPV4_VXLAN_TCP_IPV4_INDEX]; + } + if (gro_ctx->gro_types & RTE_GRO_TCP_IPV4) { + do_tcp4_gro = 1; + tcp_tbl = gro_ctx->tbls[RTE_GRO_TCP_IPV4_INDEX]; + } - tcp_tbl = gro_ctx->tbls[RTE_GRO_TCP_IPV4_INDEX]; current_time = rte_rdtsc(); for (i = 0; i < nb_pkts; i++) { - if (IS_IPV4_TCP_PKT(pkts[i]->packet_type)) { + if (do_vxlan_tcp4_gro && IS_IPV4_VXLAN_TCP4_PKT( + pkts[i]->packet_type)) { + if (gro_vxlan_tcp4_reassemble(pkts[i], vxlan_tbl, + current_time) < 0) + unprocess_pkts[unprocess_num++] = pkts[i]; + } else if (do_tcp4_gro && IS_IPV4_TCP_PKT( + pkts[i]->packet_type)) { if (gro_tcp4_reassemble(pkts[i], tcp_tbl, current_time) < 0) unprocess_pkts[unprocess_num++] = pkts[i]; @@ -235,18 +305,26 @@ rte_gro_timeout_flush(void *ctx, { struct gro_ctx *gro_ctx = ctx; uint64_t flush_timestamp; + uint16_t num = 0; gro_types = gro_types & gro_ctx->gro_types; flush_timestamp = rte_rdtsc() - timeout_cycles; - if (gro_types & RTE_GRO_TCP_IPV4) { - return gro_tcp4_tbl_timeout_flush( + if (gro_types & RTE_GRO_IPV4_VXLAN_TCP_IPV4) { + num = gro_vxlan_tcp4_tbl_timeout_flush(gro_ctx->tbls[ + RTE_GRO_IPV4_VXLAN_TCP_IPV4_INDEX], + flush_timestamp, out, max_nb_out); + } + + max_nb_out -= num; + if ((gro_types & RTE_GRO_TCP_IPV4) && max_nb_out > 0) { + num += gro_tcp4_tbl_timeout_flush( gro_ctx->tbls[RTE_GRO_TCP_IPV4_INDEX], flush_timestamp, - out, max_nb_out); + &out[num], max_nb_out); } - return 0; + return num; } uint64_t diff --git a/lib/librte_gro/rte_gro.h b/lib/librte_gro/rte_gro.h index c28b3a2..9d676ce 100644 --- a/lib/librte_gro/rte_gro.h +++ b/lib/librte_gro/rte_gro.h @@ -57,6 +57,9 @@ extern "C" { #define RTE_GRO_TCP_IPV4_INDEX 0 #define RTE_GRO_TCP_IPV4 (1ULL << RTE_GRO_TCP_IPV4_INDEX) /**< TCP/IPv4 GRO flag */ +#define RTE_GRO_IPV4_VXLAN_TCP_IPV4_INDEX 1 +#define RTE_GRO_IPV4_VXLAN_TCP_IPV4 (1ULL << RTE_GRO_IPV4_VXLAN_TCP_IPV4_INDEX) +/**< VxLAN GRO flag. */ /** * Structure used to create GRO context objects or used to pass -- 2.7.4