From patchwork Tue Jan 26 06:20:20 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jingjing Wu X-Patchwork-Id: 10077 Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [IPv6:::1]) by dpdk.org (Postfix) with ESMTP id 5BE939215; Tue, 26 Jan 2016 07:20:52 +0100 (CET) Received: from mga14.intel.com (mga14.intel.com [192.55.52.115]) by dpdk.org (Postfix) with ESMTP id A37719214 for ; Tue, 26 Jan 2016 07:20:50 +0100 (CET) Received: from orsmga003.jf.intel.com ([10.7.209.27]) by fmsmga103.fm.intel.com with ESMTP; 25 Jan 2016 22:20:49 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.22,348,1449561600"; d="scan'208";a="734316317" Received: from shvmail01.sh.intel.com ([10.239.29.42]) by orsmga003.jf.intel.com with ESMTP; 25 Jan 2016 22:20:48 -0800 Received: from shecgisg004.sh.intel.com (shecgisg004.sh.intel.com [10.239.29.89]) by shvmail01.sh.intel.com with ESMTP id u0Q6KkX0021471; Tue, 26 Jan 2016 14:20:46 +0800 Received: from shecgisg004.sh.intel.com (localhost [127.0.0.1]) by shecgisg004.sh.intel.com (8.13.6/8.13.6/SuSE Linux 0.8) with ESMTP id u0Q6KgLP005806; Tue, 26 Jan 2016 14:20:44 +0800 Received: (from wujingji@localhost) by shecgisg004.sh.intel.com (8.13.6/8.13.6/Submit) id u0Q6Kg4i005802; Tue, 26 Jan 2016 14:20:42 +0800 From: Jingjing Wu To: dev@dpdk.org Date: Tue, 26 Jan 2016 14:20:20 +0800 Message-Id: <1453789224-5716-9-git-send-email-jingjing.wu@intel.com> X-Mailer: git-send-email 1.7.4.1 In-Reply-To: <1453789224-5716-1-git-send-email-jingjing.wu@intel.com> References: <1453789224-5716-1-git-send-email-jingjing.wu@intel.com> Subject: [dpdk-dev] [PATCH 08/12] i40e: extend flow director to filter by tunnel ID X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" This patch extended flow director to select Vxlan/GRE tunnel ID as filter's input set and program the filter rule with the defined tunnel type. Signed-off-by: Jingjing Wu --- drivers/net/i40e/i40e_ethdev.c | 11 ++++ drivers/net/i40e/i40e_fdir.c | 125 ++++++++++++++++++++++++++++++----------- 2 files changed, 102 insertions(+), 34 deletions(-) diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c index 32ffc9f..62cdf81 100644 --- a/drivers/net/i40e/i40e_ethdev.c +++ b/drivers/net/i40e/i40e_ethdev.c @@ -6555,48 +6555,59 @@ i40e_get_valid_input_set(enum i40e_filter_pctype pctype, */ static const uint64_t valid_fdir_inset_table[] = { [I40E_FILTER_PCTYPE_FRAG_IPV4] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV4_SRC | I40E_INSET_IPV4_DST | I40E_INSET_IPV4_TOS | I40E_INSET_IPV4_PROTO | I40E_INSET_IPV4_TTL, [I40E_FILTER_PCTYPE_NONF_IPV4_UDP] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV4_SRC | I40E_INSET_IPV4_DST | I40E_INSET_IPV4_TOS | I40E_INSET_IPV4_TTL | I40E_INSET_SRC_PORT | I40E_INSET_DST_PORT, [I40E_FILTER_PCTYPE_NONF_IPV4_TCP] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV4_SRC | I40E_INSET_IPV4_DST | I40E_INSET_IPV4_TOS | I40E_INSET_IPV4_TTL | I40E_INSET_SRC_PORT | I40E_INSET_DST_PORT, [I40E_FILTER_PCTYPE_NONF_IPV4_SCTP] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV4_SRC | I40E_INSET_IPV4_DST | I40E_INSET_IPV4_TOS | I40E_INSET_IPV4_TTL | I40E_INSET_SRC_PORT | I40E_INSET_DST_PORT | I40E_INSET_SCTP_VT, [I40E_FILTER_PCTYPE_NONF_IPV4_OTHER] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV4_SRC | I40E_INSET_IPV4_DST | I40E_INSET_IPV4_TOS | I40E_INSET_IPV4_PROTO | I40E_INSET_IPV4_TTL, [I40E_FILTER_PCTYPE_FRAG_IPV6] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV6_SRC | I40E_INSET_IPV6_DST | I40E_INSET_IPV6_TC | I40E_INSET_IPV6_NEXT_HDR | I40E_INSET_IPV6_HOP_LIMIT, [I40E_FILTER_PCTYPE_NONF_IPV6_UDP] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV6_SRC | I40E_INSET_IPV6_DST | I40E_INSET_IPV6_TC | I40E_INSET_IPV6_HOP_LIMIT | I40E_INSET_SRC_PORT | I40E_INSET_DST_PORT, [I40E_FILTER_PCTYPE_NONF_IPV6_TCP] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV6_SRC | I40E_INSET_IPV6_DST | I40E_INSET_IPV6_TC | I40E_INSET_IPV6_HOP_LIMIT | I40E_INSET_SRC_PORT | I40E_INSET_DST_PORT, [I40E_FILTER_PCTYPE_NONF_IPV6_SCTP] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV6_SRC | I40E_INSET_IPV6_DST | I40E_INSET_IPV6_TC | I40E_INSET_IPV6_HOP_LIMIT | I40E_INSET_SRC_PORT | I40E_INSET_DST_PORT | I40E_INSET_SCTP_VT, [I40E_FILTER_PCTYPE_NONF_IPV6_OTHER] = + I40E_INSET_TUNNEL_ID | I40E_INSET_IPV6_SRC | I40E_INSET_IPV6_DST | I40E_INSET_IPV6_TC | I40E_INSET_IPV6_NEXT_HDR | I40E_INSET_IPV6_HOP_LIMIT, [I40E_FILTER_PCTYPE_L2_PAYLOAD] = + I40E_INSET_TUNNEL_ID | I40E_INSET_LAST_ETHER_TYPE, }; diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c index 5ea97e5..7566017 100644 --- a/drivers/net/i40e/i40e_fdir.c +++ b/drivers/net/i40e/i40e_fdir.c @@ -688,11 +688,13 @@ i40e_fdir_configure(struct rte_eth_dev *dev) static inline void i40e_fdir_fill_eth_ip_head(const struct rte_eth_fdir_input *fdir_input, - unsigned char *raw_pkt) + unsigned char *pkt, bool need_mac) { - struct ether_hdr *ether = (struct ether_hdr *)raw_pkt; - struct ipv4_hdr *ip; - struct ipv6_hdr *ip6; + struct ether_hdr *ether = (struct ether_hdr *)pkt; + struct ipv4_hdr *ip = + (struct ipv4_hdr *)(pkt + sizeof(struct ether_hdr)); + struct ipv6_hdr *ip6 = + (struct ipv6_hdr *)(pkt + sizeof(struct ether_hdr)); static const uint8_t next_proto[] = { [RTE_ETH_FLOW_FRAG_IPV4] = IPPROTO_IP, [RTE_ETH_FLOW_NONFRAG_IPV4_TCP] = IPPROTO_TCP, @@ -708,16 +710,18 @@ i40e_fdir_fill_eth_ip_head(const struct rte_eth_fdir_input *fdir_input, switch (fdir_input->flow_type) { case RTE_ETH_FLOW_L2_PAYLOAD: - ether->ether_type = fdir_input->flow.l2_flow.ether_type; + if (need_mac) + ether->ether_type = fdir_input->flow.l2_flow.ether_type; break; case RTE_ETH_FLOW_NONFRAG_IPV4_TCP: case RTE_ETH_FLOW_NONFRAG_IPV4_UDP: case RTE_ETH_FLOW_NONFRAG_IPV4_SCTP: case RTE_ETH_FLOW_NONFRAG_IPV4_OTHER: case RTE_ETH_FLOW_FRAG_IPV4: - ip = (struct ipv4_hdr *)(raw_pkt + sizeof(struct ether_hdr)); - - ether->ether_type = rte_cpu_to_be_16(ETHER_TYPE_IPv4); + if (!need_mac) + ip = (struct ipv4_hdr *)pkt; + else + ether->ether_type = rte_cpu_to_be_16(ETHER_TYPE_IPv4); ip->version_ihl = I40E_FDIR_IP_DEFAULT_VERSION_IHL; /* set len to by default */ ip->total_length = rte_cpu_to_be_16(I40E_FDIR_IP_DEFAULT_LEN); @@ -741,9 +745,10 @@ i40e_fdir_fill_eth_ip_head(const struct rte_eth_fdir_input *fdir_input, case RTE_ETH_FLOW_NONFRAG_IPV6_SCTP: case RTE_ETH_FLOW_NONFRAG_IPV6_OTHER: case RTE_ETH_FLOW_FRAG_IPV6: - ip6 = (struct ipv6_hdr *)(raw_pkt + sizeof(struct ether_hdr)); - - ether->ether_type = rte_cpu_to_be_16(ETHER_TYPE_IPv6); + if (!need_mac) + ip6 = (struct ipv6_hdr *)pkt; + else + ether->ether_type = rte_cpu_to_be_16(ETHER_TYPE_IPv6); ip6->vtc_flow = rte_cpu_to_be_32(I40E_FDIR_IPv6_DEFAULT_VTC_FLOW | (fdir_input->flow.ipv6_flow.tc << @@ -775,7 +780,7 @@ i40e_fdir_fill_eth_ip_head(const struct rte_eth_fdir_input *fdir_input, } } - +#define I40E_TUNNEL_KEY_LEN 4 /* * i40e_fdir_construct_pkt - construct packet based on fields in input * @pf: board private structure @@ -787,21 +792,80 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf, const struct rte_eth_fdir_input *fdir_input, unsigned char *raw_pkt) { - unsigned char *payload, *ptr; + unsigned char *payload, *ptr, *inner_pkt = raw_pkt; struct udp_hdr *udp; struct tcp_hdr *tcp; struct sctp_hdr *sctp; uint8_t size, dst = 0; uint8_t i, pit_idx, set_idx = I40E_FLXPLD_L4_IDX; /* use l4 by default*/ - - /* fill the ethernet and IP head */ - i40e_fdir_fill_eth_ip_head(fdir_input, raw_pkt); - + bool need_mac = TRUE; + static uint8_t gre4_frame[] = {0x08, 0, + 0x45, 0, 0, 0x3A, 0, 0, 0, 0, 0x40, 0x2F, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x20, 0, 0x08, 0, + 0, 0, 0, 0}; /* Mac+ IP + GRE hdr + key */ + static uint8_t gre6_frame[] = {0x08, 0, + 0x45, 0, 0, 0x3A, 0, 0, 0, 0, 0x40, 0x2F, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x20, 0, 0x86, 0xDD, + 0, 0, 0, 0, 0}; /* Mac+ IP + GRE hdr + key */ + static uint8_t vxlan_frame[] = {0x08, 0, + 0x45, 0, 0, 0x3A, 0, 0, 0, 0, 0x40, 0x11, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0x12, 0xB5, 0x12, 0xB5, 0x08, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0}; /* Mac + IP + UDP + VXLAN hdr */ + static uint8_t nvgre_frame[] = {0x08, 0, + 0x45, 0, 0, 0x3A, 0, 0, 0, 0, 0x40, 0x2F, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x20, 0, 0x65, 0x58, + 0, 0, 0, 0}; /*Mac + IP + NVGRE hdr*/ + + inner_pkt += 2 * sizeof(struct ether_addr); + /* fill the tunnel header if required */ + switch (fdir_input->flow.tunnel_flow.tunnel_type) { + case RTE_FDIR_TUNNEL_TYPE_GRE: + if (fdir_input->flow_type == RTE_ETH_FLOW_L2_PAYLOAD) { + PMD_DRV_LOG(ERR, "GRE's inner pkt shouldn't" + " be L2 frame."); + return -EINVAL; + } + if ((fdir_input->flow_type) == RTE_ETH_FLOW_FRAG_IPV4 || + (fdir_input->flow_type) == RTE_ETH_FLOW_NONFRAG_IPV4_TCP || + (fdir_input->flow_type) == RTE_ETH_FLOW_NONFRAG_IPV4_UDP || + (fdir_input->flow_type) == RTE_ETH_FLOW_NONFRAG_IPV4_SCTP || + (fdir_input->flow_type) == RTE_ETH_FLOW_NONFRAG_IPV4_OTHER) + rte_memcpy(inner_pkt, gre4_frame, sizeof(gre4_frame)); + else + rte_memcpy(inner_pkt, gre6_frame, sizeof(gre6_frame)); + /* gre4_frame and gre4_frame have the same size */ + inner_pkt += sizeof(gre4_frame); + rte_memcpy(inner_pkt - I40E_TUNNEL_KEY_LEN, + &fdir_input->flow.tunnel_flow.tunnel_id, + I40E_TUNNEL_KEY_LEN); + need_mac = FALSE; + break; + case RTE_FDIR_TUNNEL_TYPE_VXLAN: + rte_memcpy(inner_pkt, vxlan_frame, sizeof(vxlan_frame)); + inner_pkt += sizeof(vxlan_frame); + rte_memcpy(inner_pkt - I40E_TUNNEL_KEY_LEN, + &fdir_input->flow.tunnel_flow.tunnel_id, + I40E_TUNNEL_KEY_LEN); + break; + case RTE_FDIR_TUNNEL_TYPE_NVGRE: + rte_memcpy(inner_pkt, nvgre_frame, sizeof(nvgre_frame)); + inner_pkt += sizeof(nvgre_frame); + rte_memcpy(inner_pkt - I40E_TUNNEL_KEY_LEN, + &fdir_input->flow.tunnel_flow.tunnel_id, + I40E_TUNNEL_KEY_LEN); + break; + default: + break; + } + /* fill the ethernet and IP head of inner or single frame */ + i40e_fdir_fill_eth_ip_head(fdir_input, inner_pkt, need_mac); + if (need_mac) + inner_pkt += sizeof(struct ether_hdr); /* fill the L4 head */ switch (fdir_input->flow_type) { case RTE_ETH_FLOW_NONFRAG_IPV4_UDP: - udp = (struct udp_hdr *)(raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv4_hdr)); + udp = (struct udp_hdr *)(inner_pkt + sizeof(struct ipv4_hdr)); payload = (unsigned char *)udp + sizeof(struct udp_hdr); /* * The source and destination fields in the transmitted packet @@ -814,8 +878,7 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf, break; case RTE_ETH_FLOW_NONFRAG_IPV4_TCP: - tcp = (struct tcp_hdr *)(raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv4_hdr)); + tcp = (struct tcp_hdr *)(inner_pkt + sizeof(struct ipv4_hdr)); payload = (unsigned char *)tcp + sizeof(struct tcp_hdr); /* * The source and destination fields in the transmitted packet @@ -828,8 +891,7 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf, break; case RTE_ETH_FLOW_NONFRAG_IPV4_SCTP: - sctp = (struct sctp_hdr *)(raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv4_hdr)); + sctp = (struct sctp_hdr *)(inner_pkt + sizeof(struct ipv4_hdr)); payload = (unsigned char *)sctp + sizeof(struct sctp_hdr); /* * The source and destination fields in the transmitted packet @@ -843,14 +905,12 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf, case RTE_ETH_FLOW_NONFRAG_IPV4_OTHER: case RTE_ETH_FLOW_FRAG_IPV4: - payload = raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv4_hdr); + payload = inner_pkt + sizeof(struct ipv4_hdr); set_idx = I40E_FLXPLD_L3_IDX; break; case RTE_ETH_FLOW_NONFRAG_IPV6_UDP: - udp = (struct udp_hdr *)(raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv6_hdr)); + udp = (struct udp_hdr *)(inner_pkt + sizeof(struct ipv6_hdr)); payload = (unsigned char *)udp + sizeof(struct udp_hdr); /* * The source and destination fields in the transmitted packet @@ -863,8 +923,7 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf, break; case RTE_ETH_FLOW_NONFRAG_IPV6_TCP: - tcp = (struct tcp_hdr *)(raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv6_hdr)); + tcp = (struct tcp_hdr *)(inner_pkt + sizeof(struct ipv6_hdr)); payload = (unsigned char *)tcp + sizeof(struct tcp_hdr); /* * The source and destination fields in the transmitted packet @@ -877,8 +936,7 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf, break; case RTE_ETH_FLOW_NONFRAG_IPV6_SCTP: - sctp = (struct sctp_hdr *)(raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv6_hdr)); + sctp = (struct sctp_hdr *)(inner_pkt + sizeof(struct ipv6_hdr)); payload = (unsigned char *)sctp + sizeof(struct sctp_hdr); /* * The source and destination fields in the transmitted packet @@ -892,12 +950,11 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf, case RTE_ETH_FLOW_NONFRAG_IPV6_OTHER: case RTE_ETH_FLOW_FRAG_IPV6: - payload = raw_pkt + sizeof(struct ether_hdr) + - sizeof(struct ipv6_hdr); + payload = inner_pkt + sizeof(struct ipv6_hdr); set_idx = I40E_FLXPLD_L3_IDX; break; case RTE_ETH_FLOW_L2_PAYLOAD: - payload = raw_pkt + sizeof(struct ether_hdr); + payload = inner_pkt; /* * ARP packet is a special case on which the payload * starts after the whole ARP header