From patchwork Mon Dec 21 03:50:46 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: yang_y_yi X-Patchwork-Id: 85575 X-Patchwork-Delegate: thomas@monjalon.net Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id 04BDCA09FE; Mon, 21 Dec 2020 04:52:00 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id A48AACBD5; Mon, 21 Dec 2020 04:51:04 +0100 (CET) Received: from mail-m972.mail.163.com (mail-m972.mail.163.com [123.126.97.2]) by dpdk.org (Postfix) with ESMTP id B34B9CACE for ; Mon, 21 Dec 2020 04:50:57 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=163.com; s=s110527; h=From:Subject:Date:Message-Id:MIME-Version; bh=gLyuV BBm6loJ4YsRN++WaCat+TJ7iIORIVNV7wpta0c=; b=P86ypse7LOT9QQEUdTyfL Rn6e+m/QkJ/FnQrN3/5zCQB/vJOuDdme4RzN6O/HjzwjEVarkHtBh8WQeSYvYBXW vS6XJbbJHQRWJYZrItDqzXvuFE/ZPSc9TCnO9WHeTfyEdzpex2qxiaEGSdKrvM94 Edt6jBIn3MNq/h95zHJsM8= Received: from yangyi0100.home.langchao.com (unknown [111.207.123.53]) by smtp2 (Coremail) with SMTP id GtxpCgD3vb6cG+BfUOf7Fw--.755S4; Mon, 21 Dec 2020 11:50:54 +0800 (CST) From: yang_y_yi@163.com To: dev@dpdk.org Cc: jiayu.hu@intel.com, konstantin.ananyev@intel.com, thomas@monjalon.net, yangyi01@inspur.com, yang_y_yi@163.com Date: Mon, 21 Dec 2020 11:50:46 +0800 Message-Id: <20201221035052.128292-3-yang_y_yi@163.com> X-Mailer: git-send-email 2.19.2.windows.1 In-Reply-To: <20201221035052.128292-1-yang_y_yi@163.com> References: <20201221035052.128292-1-yang_y_yi@163.com> MIME-Version: 1.0 X-CM-TRANSID: GtxpCgD3vb6cG+BfUOf7Fw--.755S4 X-Coremail-Antispam: 1Uf129KBjvAXoWfuF45GFWxJFWrZFW8Kw17ZFb_yoW5Cr15Ao WxXws0gr109r15tr4jyr1I9r4jgw10g3ZrArs8KrW8XwnFy3yDGrZ3Ww4fAa17Xr43ZFn5 u3s7Gay7XF4Iyrs3n29KB7ZKAUJUUUUU529EdanIXcx71UUUUU7v73VFW2AGmfu7bjvjm3 AaLaJ3UbIYCTnIWIevJa73UjIFyTuYvjxU7WlkDUUUU X-Originating-IP: [111.207.123.53] X-CM-SenderInfo: 51dqwsp1b1xqqrwthudrp/1tbiFh4Ci144OkMuxAAAs0 Subject: [dpdk-dev] [PATCH v1 2/8] gro: support IPv4 VXLAN TCP/IPv6 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: , Errors-To: dev-bounces@dpdk.org Sender: "dev" From: Yi Yang IPv4 VXLAN TCP/IPv6 GRO can help improve TCP/IPv6 performance in IPv4 VXLAN use case. With this enabled in DPDK, OVS DPDK can leverage it to improve VM-to-VM TCP/IPv6 performance, it will merge small adjacent IPv4 VXLAN TCP/IPv6 packets to a big IPv4 VXLAN TCP/IPv6 packet immediate after they are received from a physical NIC. It is very helpful in OVS DPDK IPv4 VXLAN use case. Signed-off-by: Yi Yang --- .../prog_guide/generic_receive_offload_lib.rst | 6 +- doc/guides/rel_notes/release_21_02.rst | 5 + lib/librte_gro/gro_vxlan_tcp6.c | 496 +++++++++++++++++++++ lib/librte_gro/gro_vxlan_tcp6.h | 156 +++++++ lib/librte_gro/meson.build | 2 +- lib/librte_gro/rte_gro.c | 81 +++- lib/librte_gro/rte_gro.h | 3 + 7 files changed, 736 insertions(+), 13 deletions(-) create mode 100644 lib/librte_gro/gro_vxlan_tcp6.c create mode 100644 lib/librte_gro/gro_vxlan_tcp6.h diff --git a/doc/guides/prog_guide/generic_receive_offload_lib.rst b/doc/guides/prog_guide/generic_receive_offload_lib.rst index fee856e..ecb5909 100644 --- a/doc/guides/prog_guide/generic_receive_offload_lib.rst +++ b/doc/guides/prog_guide/generic_receive_offload_lib.rst @@ -32,7 +32,7 @@ fragmentation is possible (i.e., DF==0). Additionally, it complies RFC Currently, the GRO library provides GRO supports for TCP/IPv4, UDP/IPv4, and TCP/IPv6 packets as well as VxLAN packets which contain an outer IPv4 -header and an inner TCP/IPv4 or UDP/IPv4 packet. +header and an inner TCP/IPv4, UDP/IPv4 or TCP/IPv6 packet. Two Sets of API --------------- @@ -161,8 +161,8 @@ Header fields deciding if two packets are neighbors include: - IPv4 ID. The IPv4 ID fields of the packets, whose DF bit is 0, should be increased by 1. -VxLAN GRO ---------- +IPv4 VxLAN TCP/IPv4 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 diff --git a/doc/guides/rel_notes/release_21_02.rst b/doc/guides/rel_notes/release_21_02.rst index 1a1aeff..035d774 100644 --- a/doc/guides/rel_notes/release_21_02.rst +++ b/doc/guides/rel_notes/release_21_02.rst @@ -60,6 +60,11 @@ New Features Added TCP/IPv6 GRO support for non-VXLAN packets, this enabled TCP/IPv6 GRO support for VLAN use case. +* **Added TCP/IPv6 GRO support for IPv4 VXLAN packets.** + + Added TCP/IPv6 GRO support for IPv4 VXLAN packets, this enabled TCP/IPv6 + GRO support for IPv4 VXLAN use case. + Removed Items ------------- diff --git a/lib/librte_gro/gro_vxlan_tcp6.c b/lib/librte_gro/gro_vxlan_tcp6.c new file mode 100644 index 0000000..d49a2dd --- /dev/null +++ b/lib/librte_gro/gro_vxlan_tcp6.c @@ -0,0 +1,496 @@ +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright(c) 2021 Inspur Corporation + */ + +#include +#include +#include +#include +#include + +#include "gro_vxlan_tcp6.h" + +void * +gro_vxlan_tcp6_tbl_create(uint16_t socket_id, + uint16_t max_flow_num, + uint16_t max_item_per_flow) +{ + struct gro_vxlan_tcp6_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_TCP6_TBL_MAX_ITEM_NUM); + + if (entries_num == 0) + return NULL; + + tbl = rte_zmalloc_socket(__func__, + sizeof(struct gro_vxlan_tcp6_tbl), + RTE_CACHE_LINE_SIZE, + socket_id); + if (tbl == NULL) + return NULL; + + size = sizeof(struct gro_vxlan_tcp6_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_tcp6_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_tcp6_tbl_destroy(void *tbl) +{ + struct gro_vxlan_tcp6_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_tcp6_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_tcp6_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_tcp6_tbl *tbl, + struct rte_mbuf *pkt, + uint64_t start_time, + uint32_t prev_idx, + uint32_t sent_seq, + uint16_t outer_ip_id, + uint8_t outer_is_atomic) +{ + uint32_t item_idx; + + item_idx = find_an_empty_item(tbl); + if (unlikely(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.nb_merged = 1; + 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_tcp6_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_tcp6_tbl *tbl, + struct vxlan_tcp6_flow_key *src, + uint32_t item_idx) +{ + struct vxlan_tcp6_flow_key *dst; + uint32_t flow_idx; + + flow_idx = find_an_empty_flow(tbl); + if (unlikely(flow_idx == INVALID_ARRAY_INDEX)) + return INVALID_ARRAY_INDEX; + + dst = &(tbl->flows[flow_idx].key); + + rte_ether_addr_copy(&(src->inner_key.eth_saddr), + &(dst->inner_key.eth_saddr)); + rte_ether_addr_copy(&(src->inner_key.eth_daddr), + &(dst->inner_key.eth_daddr)); + gro_ipv6_addr_copy(src->inner_key.ip_saddr, dst->inner_key.ip_saddr); + gro_ipv6_addr_copy(src->inner_key.ip_daddr, dst->inner_key.ip_daddr); + 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.vx_flags = src->vxlan_hdr.vx_flags; + dst->vxlan_hdr.vx_vni = src->vxlan_hdr.vx_vni; + rte_ether_addr_copy(&(src->outer_eth_saddr), &(dst->outer_eth_saddr)); + rte_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_tcp6_flow(struct vxlan_tcp6_flow_key k1, + struct vxlan_tcp6_flow_key k2) +{ + return (rte_is_same_ether_addr(&k1.outer_eth_saddr, + &k2.outer_eth_saddr) && + rte_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_tcp6_flow(k1.inner_key, k2.inner_key)); +} + +static inline int +check_vxlan_seq_option_v6(struct gro_vxlan_tcp6_item *item, + struct rte_tcp_hdr *tcp_hdr, + uint32_t sent_seq, + uint16_t outer_ip_id, + uint16_t tcp_hl, + uint16_t tcp_dl, + uint8_t outer_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 (unlikely(item->outer_is_atomic ^ outer_is_atomic)) + return 0; + + l2_offset = pkt->outer_l2_len + pkt->outer_l3_len; + cmp = check_seq_option_v6(&item->inner_item, tcp_hdr, sent_seq, + tcp_hl, tcp_dl, l2_offset); + if ((cmp > 0) && (outer_is_atomic || + (outer_ip_id == item->outer_ip_id + 1))) + /* Append the new packet. */ + return 1; + else if ((cmp < 0) && (outer_is_atomic || + (outer_ip_id + item->inner_item.nb_merged == + item->outer_ip_id))) + /* Prepend the new packet. */ + return -1; + + return 0; +} + +static inline int +merge_two_vxlan_tcp6_packets(struct gro_vxlan_tcp6_item *item, + struct rte_mbuf *pkt, + int cmp, + uint32_t sent_seq, + uint16_t outer_ip_id) +{ + if (merge_two_tcp6_packets(&item->inner_item, pkt, cmp, sent_seq, + pkt->outer_l2_len + + pkt->outer_l3_len)) { + /* Update the outer IPv4 ID to the large value. */ + item->outer_ip_id = cmp > 0 ? outer_ip_id : item->outer_ip_id; + return 1; + } + + return 0; +} + +static inline void +update_vxlan_header(struct gro_vxlan_tcp6_item *item) +{ + struct rte_ipv4_hdr *ipv4_hdr; + struct rte_ipv6_hdr *ipv6_hdr; + struct rte_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 rte_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 rte_udp_hdr *)((char *)ipv4_hdr + pkt->outer_l3_len); + udp_hdr->dgram_len = rte_cpu_to_be_16(len); + + /* Update the inner IPv6 header. */ + len -= pkt->l2_len; + ipv6_hdr = (struct rte_ipv6_hdr *)((char *)udp_hdr + pkt->l2_len); + /* IPv6 payload_len does not include IPv6 header. */ + ipv6_hdr->payload_len = rte_cpu_to_be_16(len - pkt->l3_len); +} + +int32_t +gro_vxlan_tcp6_reassemble(struct rte_mbuf *pkt, + struct gro_vxlan_tcp6_tbl *tbl, + uint64_t start_time) +{ + struct rte_ether_hdr *outer_eth_hdr, *eth_hdr; + struct rte_ipv4_hdr *outer_ipv4_hdr; + struct rte_ipv6_hdr *ipv6_hdr; + struct rte_tcp_hdr *tcp_hdr; + struct rte_udp_hdr *udp_hdr; + struct rte_vxlan_hdr *vxlan_hdr; + uint32_t sent_seq; + int32_t tcp_dl; + uint16_t frag_off, outer_ip_id; + uint8_t outer_is_atomic; + + struct vxlan_tcp6_flow_key key; + uint32_t cur_idx, prev_idx, item_idx; + uint32_t i, max_flow_num, remaining_flow_num; + int cmp; + uint16_t hdr_len; + uint8_t find; + + /* + * Don't process the packet whose TCP header length is greater + * than 60 bytes or less than 20 bytes. + */ + if (unlikely(INVALID_TCP_HDRLEN(pkt->l4_len))) + return -1; + + outer_eth_hdr = rte_pktmbuf_mtod(pkt, struct rte_ether_hdr *); + outer_ipv4_hdr = (struct rte_ipv4_hdr *)((char *)outer_eth_hdr + + pkt->outer_l2_len); + udp_hdr = (struct rte_udp_hdr *)((char *)outer_ipv4_hdr + + pkt->outer_l3_len); + vxlan_hdr = (struct rte_vxlan_hdr *)((char *)udp_hdr + + sizeof(struct rte_udp_hdr)); + eth_hdr = (struct rte_ether_hdr *)((char *)vxlan_hdr + + sizeof(struct rte_vxlan_hdr)); + ipv6_hdr = (struct rte_ipv6_hdr *)((char *)udp_hdr + pkt->l2_len); + tcp_hdr = (struct rte_tcp_hdr *)((char *)ipv6_hdr + pkt->l3_len); + + /* + * Don't process the packet which has FIN, SYN, RST, PSH, URG, + * ECE or CWR set. + */ + if (tcp_hdr->tcp_flags != RTE_TCP_ACK_FLAG) + return -1; + + hdr_len = pkt->outer_l2_len + pkt->outer_l3_len + pkt->l2_len + + pkt->l3_len + pkt->l4_len; + /* + * Don't process the packet whose payload length is less than or + * equal to 0. + */ + if (pkt->pkt_len <= hdr_len) + return -1; + tcp_dl = pkt->pkt_len - hdr_len; + + /* + * Save IPv4 ID for the packet whose DF bit is 0. For the packet + * whose DF bit is 1, IPv4 ID is ignored. + */ + frag_off = rte_be_to_cpu_16(outer_ipv4_hdr->fragment_offset); + outer_is_atomic = + (frag_off & RTE_IPV4_HDR_DF_FLAG) == RTE_IPV4_HDR_DF_FLAG; + outer_ip_id = outer_is_atomic ? 0 : + rte_be_to_cpu_16(outer_ipv4_hdr->packet_id); + + sent_seq = rte_be_to_cpu_32(tcp_hdr->sent_seq); + + rte_ether_addr_copy(&(eth_hdr->s_addr), &(key.inner_key.eth_saddr)); + rte_ether_addr_copy(&(eth_hdr->d_addr), &(key.inner_key.eth_daddr)); + gro_ipv6_addr_copy(ipv6_hdr->src_addr, key.inner_key.ip_saddr); + gro_ipv6_addr_copy(ipv6_hdr->dst_addr, key.inner_key.ip_daddr); + key.inner_key.recv_ack = tcp_hdr->recv_ack; + key.inner_key.src_port = tcp_hdr->src_port; + key.inner_key.dst_port = tcp_hdr->dst_port; + + key.vxlan_hdr.vx_flags = vxlan_hdr->vx_flags; + key.vxlan_hdr.vx_vni = vxlan_hdr->vx_vni; + rte_ether_addr_copy(&(outer_eth_hdr->s_addr), &(key.outer_eth_saddr)); + rte_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; + + /* Search for a matched flow. */ + max_flow_num = tbl->max_flow_num; + remaining_flow_num = tbl->flow_num; + find = 0; + for (i = 0; i < max_flow_num && remaining_flow_num; i++) { + if (tbl->flows[i].start_index != INVALID_ARRAY_INDEX) { + if (is_same_vxlan_tcp6_flow(tbl->flows[i].key, key)) { + find = 1; + break; + } + remaining_flow_num--; + } + } + + /* + * Can't find a matched flow. Insert a new flow and store the + * packet into the flow. + */ + if (find == 0) { + item_idx = insert_new_item(tbl, pkt, start_time, + INVALID_ARRAY_INDEX, sent_seq, outer_ip_id, + outer_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_v6(&(tbl->items[cur_idx]), tcp_hdr, + sent_seq, outer_ip_id, pkt->l4_len, + tcp_dl, outer_is_atomic); + if (cmp) { + if (merge_two_vxlan_tcp6_packets(&(tbl->items[cur_idx]), + pkt, cmp, sent_seq, + outer_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, + outer_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, outer_is_atomic + ) == INVALID_ARRAY_INDEX) + return -1; + + return 0; +} + +uint16_t +gro_vxlan_tcp6_tbl_timeout_flush(struct gro_vxlan_tcp6_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_tcp6_tbl_pkt_count(void *tbl) +{ + struct gro_vxlan_tcp6_tbl *gro_tbl = tbl; + + if (gro_tbl) + return gro_tbl->item_num; + + return 0; +} diff --git a/lib/librte_gro/gro_vxlan_tcp6.h b/lib/librte_gro/gro_vxlan_tcp6.h new file mode 100644 index 0000000..87d03d8 --- /dev/null +++ b/lib/librte_gro/gro_vxlan_tcp6.h @@ -0,0 +1,156 @@ +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright(c) 2021 Inspur Corporation + */ + +#ifndef _GRO_VXLAN_TCP6_H_ +#define _GRO_VXLAN_TCP6_H_ + +#include "gro_tcp6.h" + +#define GRO_VXLAN_TCP6_TBL_MAX_ITEM_NUM (1024UL * 1024UL) + +/* Header fields representing a VxLAN flow */ +struct vxlan_tcp6_flow_key { + struct tcp6_flow_key inner_key; + struct rte_vxlan_hdr vxlan_hdr; + + struct rte_ether_addr outer_eth_saddr; + struct rte_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_tcp6_flow { + struct vxlan_tcp6_flow_key key; + /* + * The index of the first packet in the flow. INVALID_ARRAY_INDEX + * indicates an empty flow. + */ + uint32_t start_index; +}; + +struct gro_vxlan_tcp6_item { + struct gro_tcp6_item inner_item; + /* IPv4 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 an inner TCP/IPv6 packet) + * reassembly table structure + */ +struct gro_vxlan_tcp6_tbl { + /* item array */ + struct gro_vxlan_tcp6_item *items; + /* flow array */ + struct gro_vxlan_tcp6_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 an inner TCP/IPv6 packet. + * + * @param socket_id + * Socket index for allocating the 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_tcp6_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_tcp6_tbl_destroy(void *tbl); + +/** + * This function merges a VxLAN packet which has an outer IPv4 header and + * an inner TCP/IPv6 packet. It does not process the packet, whose TCP + * header has SYN, FIN, RST, PSH, CWR, ECE or URG bit set, or which + * does not have payload. + * + * This function does not check if the packet has correct checksums and + * does not 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). It returns the packet, if + * the packet has invalid parameters (e.g. SYN bit is set) or there is no + * available space in the table. + * + * @param pkt + * Packet to reassemble + * @param tbl + * Pointer pointing to the VxLAN reassembly table + * @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 or no available + * space in the table. + */ +int32_t gro_vxlan_tcp6_reassemble(struct rte_mbuf *pkt, + struct gro_vxlan_tcp6_tbl *tbl, + uint64_t start_time); + +/** + * This function flushes timeout packets in the VxLAN reassembly table, + * and without updating checksums. + * + * @param tbl + * Pointer pointing to a VxLAN 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 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_tcp6_tbl_timeout_flush(struct gro_vxlan_tcp6_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 pointing to the VxLAN reassembly table + * + * @return + * The number of packets in the table + */ +uint32_t gro_vxlan_tcp6_tbl_pkt_count(void *tbl); +#endif diff --git a/lib/librte_gro/meson.build b/lib/librte_gro/meson.build index 6143dd0..020e7e8 100644 --- a/lib/librte_gro/meson.build +++ b/lib/librte_gro/meson.build @@ -1,6 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -sources = files('rte_gro.c', 'gro_tcp4.c', 'gro_udp4.c', 'gro_vxlan_tcp4.c', 'gro_vxlan_udp4.c', 'gro_tcp6.c') +sources = files('rte_gro.c', 'gro_tcp4.c', 'gro_udp4.c', 'gro_vxlan_tcp4.c', 'gro_vxlan_udp4.c', 'gro_tcp6.c', 'gro_vxlan_tcp6.c') headers = files('rte_gro.h') deps += ['ethdev'] diff --git a/lib/librte_gro/rte_gro.c b/lib/librte_gro/rte_gro.c index 86958d5..89fed3c 100644 --- a/lib/librte_gro/rte_gro.c +++ b/lib/librte_gro/rte_gro.c @@ -12,6 +12,7 @@ #include "gro_tcp6.h" #include "gro_udp4.h" #include "gro_vxlan_tcp4.h" +#include "gro_vxlan_tcp6.h" #include "gro_vxlan_udp4.h" typedef void *(*gro_tbl_create_fn)(uint16_t socket_id, @@ -23,15 +24,16 @@ static gro_tbl_create_fn tbl_create_fn[RTE_GRO_TYPE_MAX_NUM] = { gro_tcp4_tbl_create, gro_vxlan_tcp4_tbl_create, gro_udp4_tbl_create, gro_vxlan_udp4_tbl_create, - gro_tcp6_tbl_create, NULL}; + gro_tcp6_tbl_create, gro_vxlan_tcp6_tbl_create, NULL}; static gro_tbl_destroy_fn tbl_destroy_fn[RTE_GRO_TYPE_MAX_NUM] = { gro_tcp4_tbl_destroy, gro_vxlan_tcp4_tbl_destroy, gro_udp4_tbl_destroy, gro_vxlan_udp4_tbl_destroy, - gro_tcp6_tbl_destroy, NULL}; + gro_tcp6_tbl_destroy, gro_vxlan_tcp6_tbl_destroy, NULL}; static gro_tbl_pkt_count_fn tbl_pkt_count_fn[RTE_GRO_TYPE_MAX_NUM] = { gro_tcp4_tbl_pkt_count, gro_vxlan_tcp4_tbl_pkt_count, gro_udp4_tbl_pkt_count, gro_vxlan_udp4_tbl_pkt_count, - gro_tcp6_tbl_pkt_count, NULL}; + gro_tcp6_tbl_pkt_count, gro_vxlan_tcp6_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) && \ @@ -71,6 +73,19 @@ ((ptype & RTE_PTYPE_L4_TCP) == RTE_PTYPE_L4_TCP) && \ (RTE_ETH_IS_TUNNEL_PKT(ptype) == 0)) +#define IS_IPV4_VXLAN_TCP6_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_IPV6) || \ + ((ptype & RTE_PTYPE_INNER_L3_MASK) == \ + RTE_PTYPE_INNER_L3_IPV6_EXT) || \ + ((ptype & RTE_PTYPE_INNER_L3_MASK) == \ + RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN))) + /* * GRO context structure. It keeps the table structures, which are * used to merge packets, for different GRO types. Before using @@ -176,19 +191,25 @@ struct gro_ctx { struct gro_tcp6_flow tcp6_flows[RTE_GRO_MAX_BURST_ITEM_NUM]; struct gro_tcp6_item tcp6_items[RTE_GRO_MAX_BURST_ITEM_NUM] = {{0} }; + /* Allocate a reassembly table for VXLAN TCP/IPv6 GRO */ + struct gro_vxlan_tcp6_tbl vxlan_tcp6_tbl; + struct gro_vxlan_tcp6_flow vxlan_tcp6_flows[RTE_GRO_MAX_BURST_ITEM_NUM]; + struct gro_vxlan_tcp6_item vxlan_tcp6_items[RTE_GRO_MAX_BURST_ITEM_NUM] + = {{{0}, 0, 0} }; struct rte_mbuf *unprocess_pkts[nb_pkts]; 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_tcp_gro = 0, do_udp4_gro = 0, - do_vxlan_udp_gro = 0, do_tcp6_gro = 0; + do_vxlan_udp_gro = 0, do_tcp6_gro = 0, do_vxlan_tcp6_gro = 0; if (unlikely((param->gro_types & (RTE_GRO_IPV4_VXLAN_TCP_IPV4 | RTE_GRO_TCP_IPV4 | RTE_GRO_IPV4_VXLAN_UDP_IPV4 | RTE_GRO_UDP_IPV4 | - RTE_GRO_TCP_IPV6)) == 0)) + RTE_GRO_TCP_IPV6 | + RTE_GRO_IPV4_VXLAN_TCP_IPV6)) == 0)) return nb_pkts; /* Get the maximum number of packets */ @@ -261,7 +282,18 @@ struct gro_ctx { do_tcp6_gro = 1; } - + if (param->gro_types & RTE_GRO_IPV4_VXLAN_TCP_IPV6) { + for (i = 0; i < item_num; i++) + vxlan_tcp6_flows[i].start_index = INVALID_ARRAY_INDEX; + + vxlan_tcp6_tbl.flows = vxlan_tcp6_flows; + vxlan_tcp6_tbl.items = vxlan_tcp6_items; + vxlan_tcp6_tbl.flow_num = 0; + vxlan_tcp6_tbl.item_num = 0; + vxlan_tcp6_tbl.max_flow_num = item_num; + vxlan_tcp6_tbl.max_item_num = item_num; + do_vxlan_tcp6_gro = 1; + } for (i = 0; i < nb_pkts; i++) { /* @@ -310,6 +342,15 @@ struct gro_ctx { nb_after_gro--; else if (ret < 0) unprocess_pkts[unprocess_num++] = pkts[i]; + } else if (IS_IPV4_VXLAN_TCP6_PKT(pkts[i]->packet_type) && + do_vxlan_tcp6_gro) { + ret = gro_vxlan_tcp6_reassemble(pkts[i], + &vxlan_tcp6_tbl, 0); + if (ret > 0) + /* Merge successfully */ + nb_after_gro--; + else if (ret < 0) + unprocess_pkts[unprocess_num++] = pkts[i]; } else unprocess_pkts[unprocess_num++] = pkts[i]; } @@ -344,6 +385,11 @@ struct gro_ctx { &pkts[i], nb_pkts - i); } + if (do_vxlan_tcp6_gro) { + i += gro_vxlan_tcp6_tbl_timeout_flush(&vxlan_tcp6_tbl, + 0, &pkts[i], nb_pkts - i); + } + /* Copy unprocessed packets */ if (unprocess_num > 0) { memcpy(&pkts[i], unprocess_pkts, @@ -363,17 +409,19 @@ struct gro_ctx { { struct rte_mbuf *unprocess_pkts[nb_pkts]; struct gro_ctx *gro_ctx = ctx; - void *tcp_tbl, *udp_tbl, *vxlan_tcp_tbl, *vxlan_udp_tbl, *tcp6_tbl; + void *tcp_tbl, *udp_tbl, *vxlan_tcp_tbl, *vxlan_udp_tbl, *tcp6_tbl, + *vxlan_tcp6_tbl; uint64_t current_time; uint16_t i, unprocess_num = 0; uint8_t do_tcp4_gro, do_vxlan_tcp_gro, do_udp4_gro, do_vxlan_udp_gro, - do_tcp6_gro; + do_tcp6_gro, do_vxlan_tcp6_gro; if (unlikely((gro_ctx->gro_types & (RTE_GRO_IPV4_VXLAN_TCP_IPV4 | RTE_GRO_TCP_IPV4 | RTE_GRO_IPV4_VXLAN_UDP_IPV4 | RTE_GRO_UDP_IPV4 | - RTE_GRO_TCP_IPV6)) == 0)) + RTE_GRO_TCP_IPV6 | + RTE_GRO_IPV4_VXLAN_TCP_IPV6)) == 0)) return nb_pkts; tcp_tbl = gro_ctx->tbls[RTE_GRO_TCP_IPV4_INDEX]; @@ -381,6 +429,7 @@ struct gro_ctx { udp_tbl = gro_ctx->tbls[RTE_GRO_UDP_IPV4_INDEX]; vxlan_udp_tbl = gro_ctx->tbls[RTE_GRO_IPV4_VXLAN_UDP_IPV4_INDEX]; tcp6_tbl = gro_ctx->tbls[RTE_GRO_TCP_IPV6_INDEX]; + vxlan_tcp6_tbl = gro_ctx->tbls[RTE_GRO_IPV4_VXLAN_TCP_IPV6_INDEX]; do_tcp4_gro = (gro_ctx->gro_types & RTE_GRO_TCP_IPV4) == RTE_GRO_TCP_IPV4; @@ -392,6 +441,8 @@ struct gro_ctx { RTE_GRO_IPV4_VXLAN_UDP_IPV4; do_tcp6_gro = (gro_ctx->gro_types & RTE_GRO_TCP_IPV6) == RTE_GRO_TCP_IPV6; + do_vxlan_tcp6_gro = (gro_ctx->gro_types & RTE_GRO_IPV4_VXLAN_TCP_IPV6) + == RTE_GRO_IPV4_VXLAN_TCP_IPV6; current_time = rte_rdtsc(); @@ -421,6 +472,11 @@ struct gro_ctx { if (gro_tcp6_reassemble(pkts[i], tcp6_tbl, current_time) < 0) unprocess_pkts[unprocess_num++] = pkts[i]; + } else if (IS_IPV4_VXLAN_TCP6_PKT(pkts[i]->packet_type) && + do_vxlan_tcp6_gro) { + if (gro_vxlan_tcp6_reassemble(pkts[i], vxlan_tcp6_tbl, + current_time) < 0) + unprocess_pkts[unprocess_num++] = pkts[i]; } else unprocess_pkts[unprocess_num++] = pkts[i]; } @@ -483,6 +539,13 @@ struct gro_ctx { num += gro_tcp6_tbl_timeout_flush(gro_ctx->tbls[ RTE_GRO_TCP_IPV6_INDEX], flush_timestamp, &out[num], left_nb_out); + left_nb_out = max_nb_out - num; + } + + if ((gro_types & RTE_GRO_IPV4_VXLAN_TCP_IPV6) && left_nb_out > 0) { + num += gro_vxlan_tcp6_tbl_timeout_flush(gro_ctx->tbls[ + RTE_GRO_IPV4_VXLAN_TCP_IPV6_INDEX], + flush_timestamp, &out[num], left_nb_out); } return num; diff --git a/lib/librte_gro/rte_gro.h b/lib/librte_gro/rte_gro.h index 39b57e8..41a2bec 100644 --- a/lib/librte_gro/rte_gro.h +++ b/lib/librte_gro/rte_gro.h @@ -41,6 +41,9 @@ #define RTE_GRO_TCP_IPV6_INDEX 4 #define RTE_GRO_TCP_IPV6 (1ULL << RTE_GRO_TCP_IPV6_INDEX) /**< TCP/IPv6 GRO flag */ +#define RTE_GRO_IPV4_VXLAN_TCP_IPV6_INDEX 5 +#define RTE_GRO_IPV4_VXLAN_TCP_IPV6 (1ULL << RTE_GRO_IPV4_VXLAN_TCP_IPV6_INDEX) +/**< IPv4 VxLAN TCP/IPv6 GRO flag. */ /** * Structure used to create GRO context objects or used to pass