+/* Static initializer for items. */
+#define ITEMS(...) \
+ (const enum rte_flow_item_type []){ \
+ __VA_ARGS__, RTE_FLOW_ITEM_TYPE_END, \
+ }
+
+/* Structure to generate a simple graph of layers supported by the NIC. */
+struct tap_flow_items {
+ /* Bit-mask corresponding to what is supported for this item. */
+ const void *mask;
+ const unsigned int mask_sz; /* Bit-mask size in bytes. */
+ /*
+ * Bit-mask corresponding to the default mask, if none is provided
+ * along with the item.
+ */
+ const void *default_mask;
+ /**
+ * Conversion function from rte_flow to netlink attributes.
+ *
+ * @param item
+ * rte_flow item to convert.
+ * @param data
+ * Internal structure to store the conversion.
+ *
+ * @return
+ * 0 on success, negative value otherwise.
+ */
+ int (*convert)(const struct rte_flow_item *item, void *data);
+ /** List of possible following items. */
+ const enum rte_flow_item_type *const items;
+};
+
+/* Graph of supported items and associated actions. */
+static const struct tap_flow_items tap_flow_items[] = {
+ [RTE_FLOW_ITEM_TYPE_END] = {
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_ETH),
+ },
+ [RTE_FLOW_ITEM_TYPE_ETH] = {
+ .items = ITEMS(
+ RTE_FLOW_ITEM_TYPE_VLAN,
+ RTE_FLOW_ITEM_TYPE_IPV4,
+ RTE_FLOW_ITEM_TYPE_IPV6),
+ .mask = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
+ .src.addr_bytes = "\xff\xff\xff\xff\xff\xff",
+ .type = -1,
+ },
+ .mask_sz = sizeof(struct rte_flow_item_eth),
+ .default_mask = &rte_flow_item_eth_mask,
+ .convert = tap_flow_create_eth,
+ },
+ [RTE_FLOW_ITEM_TYPE_VLAN] = {
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_IPV4,
+ RTE_FLOW_ITEM_TYPE_IPV6),
+ .mask = &(const struct rte_flow_item_vlan){
+ .tpid = -1,
+ /* DEI matching is not supported */
+#if RTE_BYTE_ORDER == RTE_LITTLE_ENDIAN
+ .tci = 0xffef,
+#else
+ .tci = 0xefff,
+#endif
+ },
+ .mask_sz = sizeof(struct rte_flow_item_vlan),
+ .default_mask = &rte_flow_item_vlan_mask,
+ .convert = tap_flow_create_vlan,
+ },
+ [RTE_FLOW_ITEM_TYPE_IPV4] = {
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_TCP),
+ .mask = &(const struct rte_flow_item_ipv4){
+ .hdr = {
+ .src_addr = -1,
+ .dst_addr = -1,
+ .next_proto_id = -1,
+ },
+ },
+ .mask_sz = sizeof(struct rte_flow_item_ipv4),
+ .default_mask = &rte_flow_item_ipv4_mask,
+ .convert = tap_flow_create_ipv4,
+ },
+ [RTE_FLOW_ITEM_TYPE_IPV6] = {
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_TCP),
+ .mask = &(const struct rte_flow_item_ipv6){
+ .hdr = {
+ .src_addr = {
+ "\xff\xff\xff\xff\xff\xff\xff\xff"
+ "\xff\xff\xff\xff\xff\xff\xff\xff",
+ },
+ .dst_addr = {
+ "\xff\xff\xff\xff\xff\xff\xff\xff"
+ "\xff\xff\xff\xff\xff\xff\xff\xff",
+ },
+ .proto = -1,
+ },
+ },
+ .mask_sz = sizeof(struct rte_flow_item_ipv6),
+ .default_mask = &rte_flow_item_ipv6_mask,
+ .convert = tap_flow_create_ipv6,
+ },
+ [RTE_FLOW_ITEM_TYPE_UDP] = {
+ .mask = &(const struct rte_flow_item_udp){
+ .hdr = {
+ .src_port = -1,
+ .dst_port = -1,
+ },
+ },
+ .mask_sz = sizeof(struct rte_flow_item_udp),
+ .default_mask = &rte_flow_item_udp_mask,
+ .convert = tap_flow_create_udp,
+ },
+ [RTE_FLOW_ITEM_TYPE_TCP] = {
+ .mask = &(const struct rte_flow_item_tcp){
+ .hdr = {
+ .src_port = -1,
+ .dst_port = -1,
+ },
+ },
+ .mask_sz = sizeof(struct rte_flow_item_tcp),
+ .default_mask = &rte_flow_item_tcp_mask,
+ .convert = tap_flow_create_tcp,
+ },
+};
+
+/*
+ * TC rules, by growing priority
+ *
+ * Remote netdevice Tap netdevice
+ * +-------------+-------------+ +-------------+-------------+
+ * | Ingress | Egress | | Ingress | Egress |
+ * |-------------|-------------| |-------------|-------------|
+ * | | \ / | | | REMOTE TX | prio 1
+ * | | \ / | | | \ / | prio 2
+ * | EXPLICIT | \ / | | EXPLICIT | \ / | .
+ * | | \ / | | | \ / | .
+ * | RULES | X | | RULES | X | .
+ * | . | / \ | | . | / \ | .
+ * | . | / \ | | . | / \ | .
+ * | . | / \ | | . | / \ | .
+ * | . | / \ | | . | / \ | .
+ *
+ * .... .... .... ....
+ *
+ * | . | \ / | | . | \ / | .
+ * | . | \ / | | . | \ / | .
+ * | | \ / | | | \ / |
+ * | LOCAL_MAC | \ / | | \ / | \ / | last prio - 5
+ * | PROMISC | X | | \ / | X | last prio - 4
+ * | ALLMULTI | / \ | | X | / \ | last prio - 3
+ * | BROADCAST | / \ | | / \ | / \ | last prio - 2
+ * | BROADCASTV6 | / \ | | / \ | / \ | last prio - 1
+ * | xx | / \ | | ISOLATE | / \ | last prio
+ * +-------------+-------------+ +-------------+-------------+
+ *
+ * The implicit flow rules are stored in a list in with mandatorily the last two
+ * being the ISOLATE and REMOTE_TX rules. e.g.:
+ *
+ * LOCAL_MAC -> BROADCAST -> BROADCASTV6 -> REMOTE_TX -> ISOLATE -> NULL
+ *
+ * That enables tap_flow_isolate() to remove implicit rules by popping the list
+ * head and remove it as long as it applies on the remote netdevice. The
+ * implicit rule for TX redirection is not removed, as isolate concerns only
+ * incoming traffic.
+ */
+
+static struct remote_rule implicit_rte_flows[TAP_REMOTE_MAX_IDX] = {
+ [TAP_REMOTE_LOCAL_MAC] = {
+ .attr = {
+ .group = MAX_GROUP,
+ .priority = PRIORITY_MASK - TAP_REMOTE_LOCAL_MAC,
+ .ingress = 1,
+ },
+ .items[0] = {
+ .type = RTE_FLOW_ITEM_TYPE_ETH,
+ .mask = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
+ },
+ },
+ .items[1] = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ .mirred = TCA_EGRESS_REDIR,
+ },
+ [TAP_REMOTE_BROADCAST] = {
+ .attr = {
+ .group = MAX_GROUP,
+ .priority = PRIORITY_MASK - TAP_REMOTE_BROADCAST,
+ .ingress = 1,
+ },
+ .items[0] = {
+ .type = RTE_FLOW_ITEM_TYPE_ETH,
+ .mask = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
+ },
+ .spec = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
+ },
+ },
+ .items[1] = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ .mirred = TCA_EGRESS_MIRROR,
+ },
+ [TAP_REMOTE_BROADCASTV6] = {
+ .attr = {
+ .group = MAX_GROUP,
+ .priority = PRIORITY_MASK - TAP_REMOTE_BROADCASTV6,
+ .ingress = 1,
+ },
+ .items[0] = {
+ .type = RTE_FLOW_ITEM_TYPE_ETH,
+ .mask = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\x33\x33\x00\x00\x00\x00",
+ },
+ .spec = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\x33\x33\x00\x00\x00\x00",
+ },
+ },
+ .items[1] = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ .mirred = TCA_EGRESS_MIRROR,
+ },
+ [TAP_REMOTE_PROMISC] = {
+ .attr = {
+ .group = MAX_GROUP,
+ .priority = PRIORITY_MASK - TAP_REMOTE_PROMISC,
+ .ingress = 1,
+ },
+ .items[0] = {
+ .type = RTE_FLOW_ITEM_TYPE_VOID,
+ },
+ .items[1] = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ .mirred = TCA_EGRESS_MIRROR,
+ },
+ [TAP_REMOTE_ALLMULTI] = {
+ .attr = {
+ .group = MAX_GROUP,
+ .priority = PRIORITY_MASK - TAP_REMOTE_ALLMULTI,
+ .ingress = 1,
+ },
+ .items[0] = {
+ .type = RTE_FLOW_ITEM_TYPE_ETH,
+ .mask = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\x01\x00\x00\x00\x00\x00",
+ },
+ .spec = &(const struct rte_flow_item_eth){
+ .dst.addr_bytes = "\x01\x00\x00\x00\x00\x00",
+ },
+ },
+ .items[1] = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ .mirred = TCA_EGRESS_MIRROR,
+ },
+ [TAP_REMOTE_TX] = {
+ .attr = {
+ .group = 0,
+ .priority = TAP_REMOTE_TX,
+ .egress = 1,
+ },
+ .items[0] = {
+ .type = RTE_FLOW_ITEM_TYPE_VOID,
+ },
+ .items[1] = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ .mirred = TCA_EGRESS_MIRROR,
+ },
+ [TAP_ISOLATE] = {
+ .attr = {
+ .group = MAX_GROUP,
+ .priority = PRIORITY_MASK - TAP_ISOLATE,
+ .ingress = 1,
+ },
+ .items[0] = {
+ .type = RTE_FLOW_ITEM_TYPE_VOID,
+ },
+ .items[1] = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ },
+};
+
+/**
+ * Make as much checks as possible on an Ethernet item, and if a flow is
+ * provided, fill it appropriately with Ethernet info.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in, out] data
+ * Additional data structure to tell next layers we've been here.
+ *
+ * @return
+ * 0 if checks are alright, -1 otherwise.
+ */
+static int
+tap_flow_create_eth(const struct rte_flow_item *item, void *data)
+{
+ struct convert_data *info = (struct convert_data *)data;
+ const struct rte_flow_item_eth *spec = item->spec;
+ const struct rte_flow_item_eth *mask = item->mask;
+ struct rte_flow *flow = info->flow;
+ struct nlmsg *msg;
+
+ /* use default mask if none provided */
+ if (!mask)
+ mask = tap_flow_items[RTE_FLOW_ITEM_TYPE_ETH].default_mask;
+ /* TC does not support eth_type masking. Only accept if exact match. */
+ if (mask->type && mask->type != 0xffff)
+ return -1;
+ if (!spec)
+ return 0;
+ /* store eth_type for consistency if ipv4/6 pattern item comes next */
+ if (spec->type & mask->type)
+ info->eth_type = spec->type;
+ if (!flow)
+ return 0;
+ msg = &flow->msg;
+ if (!is_zero_ether_addr(&spec->dst)) {
+ tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_ETH_DST, ETHER_ADDR_LEN,
+ &spec->dst.addr_bytes);
+ tap_nlattr_add(&msg->nh,
+ TCA_FLOWER_KEY_ETH_DST_MASK, ETHER_ADDR_LEN,
+ &mask->dst.addr_bytes);
+ }
+ if (!is_zero_ether_addr(&mask->src)) {
+ tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_ETH_SRC, ETHER_ADDR_LEN,
+ &spec->src.addr_bytes);
+ tap_nlattr_add(&msg->nh,
+ TCA_FLOWER_KEY_ETH_SRC_MASK, ETHER_ADDR_LEN,
+ &mask->src.addr_bytes);
+ }
+ return 0;
+}
+
+/**
+ * Make as much checks as possible on a VLAN item, and if a flow is provided,
+ * fill it appropriately with VLAN info.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in, out] data
+ * Additional data structure to tell next layers we've been here.
+ *
+ * @return
+ * 0 if checks are alright, -1 otherwise.
+ */
+static int
+tap_flow_create_vlan(const struct rte_flow_item *item, void *data)
+{
+ struct convert_data *info = (struct convert_data *)data;
+ const struct rte_flow_item_vlan *spec = item->spec;
+ const struct rte_flow_item_vlan *mask = item->mask;
+ struct rte_flow *flow = info->flow;
+ struct nlmsg *msg;
+
+ /* use default mask if none provided */
+ if (!mask)
+ mask = tap_flow_items[RTE_FLOW_ITEM_TYPE_VLAN].default_mask;
+ /* TC does not support tpid masking. Only accept if exact match. */
+ if (mask->tpid && mask->tpid != 0xffff)
+ return -1;
+ /* Double-tagging not supported. */
+ if (spec && mask->tpid && spec->tpid != htons(ETH_P_8021Q))
+ return -1;
+ info->vlan = 1;
+ if (!flow)
+ return 0;
+ msg = &flow->msg;
+ msg->t.tcm_info = TC_H_MAKE(msg->t.tcm_info, htons(ETH_P_8021Q));
+#define VLAN_PRIO(tci) ((tci) >> 13)
+#define VLAN_ID(tci) ((tci) & 0xfff)
+ if (!spec)
+ return 0;
+ if (spec->tci) {
+ uint16_t tci = ntohs(spec->tci) & mask->tci;
+ uint16_t prio = VLAN_PRIO(tci);
+ uint8_t vid = VLAN_ID(tci);
+
+ if (prio)
+ tap_nlattr_add8(&msg->nh,
+ TCA_FLOWER_KEY_VLAN_PRIO, prio);
+ if (vid)
+ tap_nlattr_add16(&msg->nh,
+ TCA_FLOWER_KEY_VLAN_ID, vid);
+ }
+ return 0;
+}
+
+/**
+ * Make as much checks as possible on an IPv4 item, and if a flow is provided,
+ * fill it appropriately with IPv4 info.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in, out] data
+ * Additional data structure to tell next layers we've been here.
+ *
+ * @return
+ * 0 if checks are alright, -1 otherwise.
+ */
+static int
+tap_flow_create_ipv4(const struct rte_flow_item *item, void *data)
+{
+ struct convert_data *info = (struct convert_data *)data;
+ const struct rte_flow_item_ipv4 *spec = item->spec;
+ const struct rte_flow_item_ipv4 *mask = item->mask;
+ struct rte_flow *flow = info->flow;
+ struct nlmsg *msg;
+
+ /* use default mask if none provided */
+ if (!mask)
+ mask = tap_flow_items[RTE_FLOW_ITEM_TYPE_IPV4].default_mask;
+ /* check that previous eth type is compatible with ipv4 */
+ if (info->eth_type && info->eth_type != htons(ETH_P_IP))
+ return -1;
+ /* store ip_proto for consistency if udp/tcp pattern item comes next */
+ if (spec)
+ info->ip_proto = spec->hdr.next_proto_id;
+ if (!flow)
+ return 0;
+ msg = &flow->msg;
+ if (!info->eth_type)
+ info->eth_type = htons(ETH_P_IP);
+ if (!spec)
+ return 0;
+ if (spec->hdr.dst_addr) {
+ tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_DST,
+ spec->hdr.dst_addr);
+ tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_DST_MASK,
+ mask->hdr.dst_addr);
+ }
+ if (spec->hdr.src_addr) {
+ tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_SRC,
+ spec->hdr.src_addr);
+ tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_SRC_MASK,
+ mask->hdr.src_addr);
+ }
+ if (spec->hdr.next_proto_id)
+ tap_nlattr_add8(&msg->nh, TCA_FLOWER_KEY_IP_PROTO,
+ spec->hdr.next_proto_id);
+ return 0;
+}
+
+/**
+ * Make as much checks as possible on an IPv6 item, and if a flow is provided,
+ * fill it appropriately with IPv6 info.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in, out] data
+ * Additional data structure to tell next layers we've been here.
+ *
+ * @return
+ * 0 if checks are alright, -1 otherwise.
+ */
+static int
+tap_flow_create_ipv6(const struct rte_flow_item *item, void *data)
+{
+ struct convert_data *info = (struct convert_data *)data;
+ const struct rte_flow_item_ipv6 *spec = item->spec;
+ const struct rte_flow_item_ipv6 *mask = item->mask;
+ struct rte_flow *flow = info->flow;
+ uint8_t empty_addr[16] = { 0 };
+ struct nlmsg *msg;
+
+ /* use default mask if none provided */
+ if (!mask)
+ mask = tap_flow_items[RTE_FLOW_ITEM_TYPE_IPV6].default_mask;
+ /* check that previous eth type is compatible with ipv6 */
+ if (info->eth_type && info->eth_type != htons(ETH_P_IPV6))
+ return -1;
+ /* store ip_proto for consistency if udp/tcp pattern item comes next */
+ if (spec)
+ info->ip_proto = spec->hdr.proto;
+ if (!flow)
+ return 0;
+ msg = &flow->msg;
+ if (!info->eth_type)
+ info->eth_type = htons(ETH_P_IPV6);
+ if (!spec)
+ return 0;
+ if (memcmp(spec->hdr.dst_addr, empty_addr, 16)) {
+ tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_DST,
+ sizeof(spec->hdr.dst_addr), &spec->hdr.dst_addr);
+ tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_DST_MASK,
+ sizeof(mask->hdr.dst_addr), &mask->hdr.dst_addr);
+ }
+ if (memcmp(spec->hdr.src_addr, empty_addr, 16)) {
+ tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_SRC,
+ sizeof(spec->hdr.src_addr), &spec->hdr.src_addr);
+ tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_SRC_MASK,
+ sizeof(mask->hdr.src_addr), &mask->hdr.src_addr);
+ }
+ if (spec->hdr.proto)
+ tap_nlattr_add8(&msg->nh,
+ TCA_FLOWER_KEY_IP_PROTO, spec->hdr.proto);
+ return 0;
+}
+
+/**
+ * Make as much checks as possible on a UDP item, and if a flow is provided,
+ * fill it appropriately with UDP info.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in, out] data
+ * Additional data structure to tell next layers we've been here.
+ *
+ * @return
+ * 0 if checks are alright, -1 otherwise.
+ */
+static int
+tap_flow_create_udp(const struct rte_flow_item *item, void *data)
+{
+ struct convert_data *info = (struct convert_data *)data;
+ const struct rte_flow_item_udp *spec = item->spec;
+ const struct rte_flow_item_udp *mask = item->mask;
+ struct rte_flow *flow = info->flow;
+ struct nlmsg *msg;
+
+ /* use default mask if none provided */
+ if (!mask)
+ mask = tap_flow_items[RTE_FLOW_ITEM_TYPE_UDP].default_mask;
+ /* check that previous ip_proto is compatible with udp */
+ if (info->ip_proto && info->ip_proto != IPPROTO_UDP)
+ return -1;
+ /* TC does not support UDP port masking. Only accept if exact match. */
+ if ((mask->hdr.src_port && mask->hdr.src_port != 0xffff) ||
+ (mask->hdr.dst_port && mask->hdr.dst_port != 0xffff))
+ return -1;
+ if (!flow)
+ return 0;
+ msg = &flow->msg;
+ tap_nlattr_add8(&msg->nh, TCA_FLOWER_KEY_IP_PROTO, IPPROTO_UDP);
+ if (!spec)
+ return 0;
+ if (spec->hdr.dst_port & mask->hdr.dst_port)
+ tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_UDP_DST,
+ spec->hdr.dst_port);
+ if (spec->hdr.src_port & mask->hdr.src_port)
+ tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_UDP_SRC,
+ spec->hdr.src_port);
+ return 0;
+}
+
+/**
+ * Make as much checks as possible on a TCP item, and if a flow is provided,
+ * fill it appropriately with TCP info.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in, out] data
+ * Additional data structure to tell next layers we've been here.
+ *
+ * @return
+ * 0 if checks are alright, -1 otherwise.
+ */
+static int
+tap_flow_create_tcp(const struct rte_flow_item *item, void *data)
+{
+ struct convert_data *info = (struct convert_data *)data;
+ const struct rte_flow_item_tcp *spec = item->spec;
+ const struct rte_flow_item_tcp *mask = item->mask;
+ struct rte_flow *flow = info->flow;
+ struct nlmsg *msg;
+
+ /* use default mask if none provided */
+ if (!mask)
+ mask = tap_flow_items[RTE_FLOW_ITEM_TYPE_TCP].default_mask;
+ /* check that previous ip_proto is compatible with tcp */
+ if (info->ip_proto && info->ip_proto != IPPROTO_TCP)
+ return -1;
+ /* TC does not support TCP port masking. Only accept if exact match. */
+ if ((mask->hdr.src_port && mask->hdr.src_port != 0xffff) ||
+ (mask->hdr.dst_port && mask->hdr.dst_port != 0xffff))
+ return -1;
+ if (!flow)
+ return 0;
+ msg = &flow->msg;
+ tap_nlattr_add8(&msg->nh, TCA_FLOWER_KEY_IP_PROTO, IPPROTO_TCP);
+ if (!spec)
+ return 0;
+ if (spec->hdr.dst_port & mask->hdr.dst_port)
+ tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_TCP_DST,
+ spec->hdr.dst_port);
+ if (spec->hdr.src_port & mask->hdr.src_port)
+ tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_TCP_SRC,
+ spec->hdr.src_port);
+ return 0;
+}
+
+/**
+ * Check support for a given item.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param size
+ * Bit-Mask size in bytes.
+ * @param[in] supported_mask
+ * Bit-mask covering supported fields to compare with spec, last and mask in
+ * \item.
+ * @param[in] default_mask
+ * Bit-mask default mask if none is provided in \item.
+ *
+ * @return
+ * 0 on success.
+ */
+static int
+tap_flow_item_validate(const struct rte_flow_item *item,
+ unsigned int size,
+ const uint8_t *supported_mask,
+ const uint8_t *default_mask)
+{
+ int ret = 0;
+
+ /* An empty layer is allowed, as long as all fields are NULL */
+ if (!item->spec && (item->mask || item->last))
+ return -1;
+ /* Is the item spec compatible with what the NIC supports? */
+ if (item->spec && !item->mask) {
+ unsigned int i;
+ const uint8_t *spec = item->spec;
+
+ for (i = 0; i < size; ++i)
+ if ((spec[i] | supported_mask[i]) != supported_mask[i])
+ return -1;
+ /* Is the default mask compatible with what the NIC supports? */
+ for (i = 0; i < size; i++)
+ if ((default_mask[i] | supported_mask[i]) !=
+ supported_mask[i])
+ return -1;
+ }
+ /* Is the item last compatible with what the NIC supports? */
+ if (item->last && !item->mask) {
+ unsigned int i;
+ const uint8_t *spec = item->last;
+
+ for (i = 0; i < size; ++i)
+ if ((spec[i] | supported_mask[i]) != supported_mask[i])
+ return -1;
+ }
+ /* Is the item mask compatible with what the NIC supports? */
+ if (item->mask) {
+ unsigned int i;
+ const uint8_t *spec = item->mask;
+
+ for (i = 0; i < size; ++i)
+ if ((spec[i] | supported_mask[i]) != supported_mask[i])
+ return -1;
+ }
+ /**
+ * Once masked, Are item spec and item last equal?
+ * TC does not support range so anything else is invalid.
+ */
+ if (item->spec && item->last) {
+ uint8_t spec[size];
+ uint8_t last[size];
+ const uint8_t *apply = default_mask;
+ unsigned int i;
+
+ if (item->mask)
+ apply = item->mask;
+ for (i = 0; i < size; ++i) {
+ spec[i] = ((const uint8_t *)item->spec)[i] & apply[i];
+ last[i] = ((const uint8_t *)item->last)[i] & apply[i];
+ }
+ ret = memcmp(spec, last, size);
+ }
+ return ret;
+}
+
+/**
+ * Configure the kernel with a TC action and its configured parameters
+ * Handled actions: "gact", "mirred", "skbedit", "bpf"
+ *
+ * @param[in] flow
+ * Pointer to rte flow containing the netlink message
+ *
+ * @param[in, out] act_index
+ * Pointer to action sequence number in the TC command
+ *
+ * @param[in] adata
+ * Pointer to struct holding the action parameters
+ *
+ * @return
+ * -1 on failure, 0 on success
+ */
+static int
+add_action(struct rte_flow *flow, size_t *act_index, struct action_data *adata)
+{
+ struct nlmsg *msg = &flow->msg;
+
+ if (tap_nlattr_nested_start(msg, (*act_index)++) < 0)
+ return -1;
+
+ tap_nlattr_add(&msg->nh, TCA_ACT_KIND,
+ strlen(adata->id) + 1, adata->id);
+ if (tap_nlattr_nested_start(msg, TCA_ACT_OPTIONS) < 0)
+ return -1;
+ if (strcmp("gact", adata->id) == 0) {
+ tap_nlattr_add(&msg->nh, TCA_GACT_PARMS, sizeof(adata->gact),
+ &adata->gact);
+ } else if (strcmp("mirred", adata->id) == 0) {
+ if (adata->mirred.eaction == TCA_EGRESS_MIRROR)
+ adata->mirred.action = TC_ACT_PIPE;
+ else /* REDIRECT */
+ adata->mirred.action = TC_ACT_STOLEN;
+ tap_nlattr_add(&msg->nh, TCA_MIRRED_PARMS,
+ sizeof(adata->mirred),
+ &adata->mirred);
+ } else if (strcmp("skbedit", adata->id) == 0) {
+ tap_nlattr_add(&msg->nh, TCA_SKBEDIT_PARMS,
+ sizeof(adata->skbedit.skbedit),
+ &adata->skbedit.skbedit);
+ tap_nlattr_add16(&msg->nh, TCA_SKBEDIT_QUEUE_MAPPING,
+ adata->skbedit.queue);
+ } else if (strcmp("bpf", adata->id) == 0) {
+ tap_nlattr_add32(&msg->nh, TCA_ACT_BPF_FD, adata->bpf.bpf_fd);
+ tap_nlattr_add(&msg->nh, TCA_ACT_BPF_NAME,
+ strlen(adata->bpf.annotation) + 1,
+ adata->bpf.annotation);
+ tap_nlattr_add(&msg->nh, TCA_ACT_BPF_PARMS,
+ sizeof(adata->bpf.bpf),
+ &adata->bpf.bpf);
+ } else {
+ return -1;
+ }
+ tap_nlattr_nested_finish(msg); /* nested TCA_ACT_OPTIONS */
+ tap_nlattr_nested_finish(msg); /* nested act_index */
+ return 0;
+}
+
+/**
+ * Helper function to send a serie of TC actions to the kernel
+ *
+ * @param[in] flow
+ * Pointer to rte flow containing the netlink message
+ *
+ * @param[in] nb_actions
+ * Number of actions in an array of action structs
+ *
+ * @param[in] data
+ * Pointer to an array of action structs
+ *
+ * @param[in] classifier_actions
+ * The classifier on behave of which the actions are configured
+ *
+ * @return
+ * -1 on failure, 0 on success
+ */
+static int
+add_actions(struct rte_flow *flow, int nb_actions, struct action_data *data,
+ int classifier_action)
+{
+ struct nlmsg *msg = &flow->msg;
+ size_t act_index = 1;
+ int i;
+
+ if (tap_nlattr_nested_start(msg, classifier_action) < 0)
+ return -1;
+ for (i = 0; i < nb_actions; i++)
+ if (add_action(flow, &act_index, data + i) < 0)
+ return -1;
+ tap_nlattr_nested_finish(msg); /* nested TCA_FLOWER_ACT */
+ return 0;
+}
+
+/**
+ * Validate a flow supported by TC.
+ * If flow param is not NULL, then also fill the netlink message inside.
+ *
+ * @param pmd
+ * Pointer to private structure.
+ * @param[in] attr
+ * Flow rule attributes.
+ * @param[in] pattern
+ * Pattern specification (list terminated by the END pattern item).
+ * @param[in] actions
+ * Associated actions (list terminated by the END action).
+ * @param[out] error
+ * Perform verbose error reporting if not NULL.
+ * @param[in, out] flow
+ * Flow structure to update.
+ * @param[in] mirred
+ * If set to TCA_EGRESS_REDIR, provided actions will be replaced with a
+ * redirection to the tap netdevice, and the TC rule will be configured
+ * on the remote netdevice in pmd.
+ * If set to TCA_EGRESS_MIRROR, provided actions will be replaced with a
+ * mirroring to the tap netdevice, and the TC rule will be configured
+ * on the remote netdevice in pmd. Matching packets will thus be duplicated.
+ * If set to 0, the standard behavior is to be used: set correct actions for
+ * the TC rule, and apply it on the tap netdevice.
+ *
+ * @return
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+static int
+priv_flow_process(struct pmd_internals *pmd,
+ const struct rte_flow_attr *attr,
+ const struct rte_flow_item items[],
+ const struct rte_flow_action actions[],
+ struct rte_flow_error *error,
+ struct rte_flow *flow,
+ int mirred)
+{
+ const struct tap_flow_items *cur_item = tap_flow_items;
+ struct convert_data data = {
+ .eth_type = 0,
+ .ip_proto = 0,
+ .flow = flow,
+ };
+ int action = 0; /* Only one action authorized for now */
+
+ if (attr->group > MAX_GROUP) {
+ rte_flow_error_set(
+ error, EINVAL, RTE_FLOW_ERROR_TYPE_ATTR_GROUP,
+ NULL, "group value too big: cannot exceed 15");
+ return -rte_errno;
+ }
+ if (attr->priority > MAX_PRIORITY) {
+ rte_flow_error_set(
+ error, EINVAL, RTE_FLOW_ERROR_TYPE_ATTR_PRIORITY,
+ NULL, "priority value too big");
+ return -rte_errno;
+ } else if (flow) {
+ uint16_t group = attr->group << GROUP_SHIFT;
+ uint16_t prio = group | (attr->priority +
+ RSS_PRIORITY_OFFSET + PRIORITY_OFFSET);
+ flow->msg.t.tcm_info = TC_H_MAKE(prio << 16,
+ flow->msg.t.tcm_info);
+ }
+ if (flow) {
+ if (mirred) {
+ /*
+ * If attr->ingress, the rule applies on remote ingress
+ * to match incoming packets
+ * If attr->egress, the rule applies on tap ingress (as
+ * seen from the kernel) to deal with packets going out
+ * from the DPDK app.
+ */
+ flow->msg.t.tcm_parent = TC_H_MAKE(TC_H_INGRESS, 0);
+ } else {
+ /* Standard rule on tap egress (kernel standpoint). */
+ flow->msg.t.tcm_parent =
+ TC_H_MAKE(MULTIQ_MAJOR_HANDLE, 0);
+ }
+ /* use flower filter type */
+ tap_nlattr_add(&flow->msg.nh, TCA_KIND, sizeof("flower"), "flower");
+ if (tap_nlattr_nested_start(&flow->msg, TCA_OPTIONS) < 0)
+ goto exit_item_not_supported;
+ }
+ for (; items->type != RTE_FLOW_ITEM_TYPE_END; ++items) {
+ const struct tap_flow_items *token = NULL;
+ unsigned int i;
+ int err = 0;
+
+ if (items->type == RTE_FLOW_ITEM_TYPE_VOID)
+ continue;
+ for (i = 0;
+ cur_item->items &&
+ cur_item->items[i] != RTE_FLOW_ITEM_TYPE_END;
+ ++i) {
+ if (cur_item->items[i] == items->type) {
+ token = &tap_flow_items[items->type];
+ break;
+ }
+ }
+ if (!token)
+ goto exit_item_not_supported;
+ cur_item = token;
+ err = tap_flow_item_validate(
+ items, cur_item->mask_sz,
+ (const uint8_t *)cur_item->mask,
+ (const uint8_t *)cur_item->default_mask);
+ if (err)
+ goto exit_item_not_supported;
+ if (flow && cur_item->convert) {
+ err = cur_item->convert(items, &data);
+ if (err)
+ goto exit_item_not_supported;
+ }
+ }
+ if (flow) {
+ if (data.vlan) {
+ tap_nlattr_add16(&flow->msg.nh, TCA_FLOWER_KEY_ETH_TYPE,
+ htons(ETH_P_8021Q));
+ tap_nlattr_add16(&flow->msg.nh,
+ TCA_FLOWER_KEY_VLAN_ETH_TYPE,
+ data.eth_type ?
+ data.eth_type : htons(ETH_P_ALL));
+ } else if (data.eth_type) {
+ tap_nlattr_add16(&flow->msg.nh, TCA_FLOWER_KEY_ETH_TYPE,
+ data.eth_type);
+ }
+ }
+ if (mirred && flow) {
+ struct action_data adata = {
+ .id = "mirred",
+ .mirred = {
+ .eaction = mirred,
+ },
+ };
+
+ /*
+ * If attr->egress && mirred, then this is a special
+ * case where the rule must be applied on the tap, to
+ * redirect packets coming from the DPDK App, out
+ * through the remote netdevice.
+ */
+ adata.mirred.ifindex = attr->ingress ? pmd->if_index :
+ pmd->remote_if_index;
+ if (mirred == TCA_EGRESS_MIRROR)
+ adata.mirred.action = TC_ACT_PIPE;
+ else
+ adata.mirred.action = TC_ACT_STOLEN;
+ if (add_actions(flow, 1, &adata, TCA_FLOWER_ACT) < 0)
+ goto exit_action_not_supported;
+ else
+ goto end;
+ }
+ for (; actions->type != RTE_FLOW_ACTION_TYPE_END; ++actions) {
+ int err = 0;
+
+ if (actions->type == RTE_FLOW_ACTION_TYPE_VOID) {
+ continue;
+ } else if (actions->type == RTE_FLOW_ACTION_TYPE_DROP) {
+ if (action)
+ goto exit_action_not_supported;
+ action = 1;
+ if (flow) {
+ struct action_data adata = {
+ .id = "gact",
+ .gact = {
+ .action = TC_ACT_SHOT,
+ },
+ };
+
+ err = add_actions(flow, 1, &adata,
+ TCA_FLOWER_ACT);
+ }
+ } else if (actions->type == RTE_FLOW_ACTION_TYPE_PASSTHRU) {
+ if (action)
+ goto exit_action_not_supported;
+ action = 1;
+ if (flow) {
+ struct action_data adata = {
+ .id = "gact",
+ .gact = {
+ /* continue */
+ .action = TC_ACT_UNSPEC,
+ },
+ };
+
+ err = add_actions(flow, 1, &adata,
+ TCA_FLOWER_ACT);
+ }
+ } else if (actions->type == RTE_FLOW_ACTION_TYPE_QUEUE) {
+ const struct rte_flow_action_queue *queue =
+ (const struct rte_flow_action_queue *)
+ actions->conf;
+
+ if (action)
+ goto exit_action_not_supported;
+ action = 1;
+ if (!queue ||
+ (queue->index > pmd->dev->data->nb_rx_queues - 1))
+ goto exit_action_not_supported;
+ if (flow) {
+ struct action_data adata = {
+ .id = "skbedit",
+ .skbedit = {
+ .skbedit = {
+ .action = TC_ACT_PIPE,
+ },
+ .queue = queue->index,
+ },
+ };
+
+ err = add_actions(flow, 1, &adata,
+ TCA_FLOWER_ACT);
+ }
+ } else if (actions->type == RTE_FLOW_ACTION_TYPE_RSS) {
+ const struct rte_flow_action_rss *rss =
+ (const struct rte_flow_action_rss *)
+ actions->conf;
+
+ if (action++)
+ goto exit_action_not_supported;
+
+ if (!pmd->rss_enabled) {
+ err = rss_enable(pmd, attr, error);
+ if (err)
+ goto exit_action_not_supported;
+ }
+ if (flow && rss)
+ err = rss_add_actions(flow, pmd, rss, error);
+ } else {
+ goto exit_action_not_supported;
+ }
+ if (err)
+ goto exit_action_not_supported;
+ }
+end:
+ if (flow)
+ tap_nlattr_nested_finish(&flow->msg); /* nested TCA_OPTIONS */
+ return 0;
+exit_item_not_supported:
+ rte_flow_error_set(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM,
+ items, "item not supported");
+ return -rte_errno;
+exit_action_not_supported:
+ rte_flow_error_set(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ACTION,
+ actions, "action not supported");
+ return -rte_errno;
+}
+
+
+