X-Git-Url: http://git.cascardo.eti.br/?a=blobdiff_plain;f=lib%2Fflow.h;h=a1c6e9751ee2d3f374ef15d370abff61c86de306;hb=06994f879c9d;hp=c834cb5540e60c1cc36d09287c68936bf2db0856;hpb=b7807e4f64c8c64bb6000767de72368306a95c90;p=cascardo%2Fovs.git diff --git a/lib/flow.h b/lib/flow.h index c834cb554..a1c6e9751 100644 --- a/lib/flow.h +++ b/lib/flow.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2008, 2009, 2010, 2011, 2012, 2013, 2014 Nicira, Inc. + * Copyright (c) 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015 Nicira, Inc. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -32,16 +32,25 @@ struct dpif_flow_stats; struct ds; struct flow_wildcards; struct minimask; -struct ofpbuf; +struct dp_packet; struct pkt_metadata; +struct match; /* This sequence number should be incremented whenever anything involving flows * or the wildcarding of flows changes. This will cause build assertion * failures in places which likely need to be updated. */ -#define FLOW_WC_SEQ 26 +#define FLOW_WC_SEQ 32 +/* Number of Open vSwitch extension 32-bit registers. */ #define FLOW_N_REGS 8 BUILD_ASSERT_DECL(FLOW_N_REGS <= NXM_NX_MAX_REGS); +BUILD_ASSERT_DECL(FLOW_N_REGS % 2 == 0); /* Even. */ + +/* Number of OpenFlow 1.5+ 64-bit registers. + * + * Each of these overlays a pair of Open vSwitch 32-bit registers, so there + * are half as many of them.*/ +#define FLOW_N_XREGS (FLOW_N_REGS / 2) /* Used for struct flow's dl_type member for frames that have no Ethernet * type, that is, pure 802.2 frames. */ @@ -58,6 +67,9 @@ BUILD_ASSERT_DECL(FLOW_NW_FRAG_LATER == NX_IP_FRAG_LATER); #define FLOW_TNL_F_DONT_FRAGMENT (1 << 0) #define FLOW_TNL_F_CSUM (1 << 1) #define FLOW_TNL_F_KEY (1 << 2) +#define FLOW_TNL_F_OAM (1 << 3) + +#define FLOW_TNL_F_MASK ((1 << 4) - 1) const char *flow_tun_flag_to_string(uint32_t flags); @@ -73,8 +85,8 @@ const char *flow_tun_flag_to_string(uint32_t flags); * * The meaning of 'in_port' is context-dependent. In most cases, it is a * 16-bit OpenFlow 1.0 port number. In the software datapath interface (dpif) - * layer and its implementations (e.g. dpif-linux, dpif-netdev), it is instead - * a 32-bit datapath port number. + * layer and its implementations (e.g. dpif-netlink, dpif-netdev), it is + * instead a 32-bit datapath port number. * * The fields are organized in four segments to facilitate staged lookup, where * lower layer fields are first used to determine if the later fields need to @@ -84,94 +96,95 @@ const char *flow_tun_flag_to_string(uint32_t flags); * reflected in miniflow_extract()! */ struct flow { - /* L1 */ + /* Metadata */ struct flow_tnl tunnel; /* Encapsulating tunnel parameters. */ ovs_be64 metadata; /* OpenFlow Metadata. */ uint32_t regs[FLOW_N_REGS]; /* Registers. */ uint32_t skb_priority; /* Packet priority for QoS. */ uint32_t pkt_mark; /* Packet mark. */ - uint32_t recirc_id; /* Must be exact match. */ + uint32_t dp_hash; /* Datapath computed hash value. The exact + * computation is opaque to the user space. */ union flow_in_port in_port; /* Input port.*/ + uint32_t recirc_id; /* Must be exact match. */ + uint32_t conj_id; /* Conjunction ID. */ + ofp_port_t actset_output; /* Output port in action set. */ + uint8_t pad1[6]; /* Pad to 64 bits. */ - /* L2, Order the same as in the Ethernet header! */ - uint8_t dl_dst[6]; /* Ethernet destination address. */ - uint8_t dl_src[6]; /* Ethernet source address. */ + /* L2, Order the same as in the Ethernet header! (64-bit aligned) */ + uint8_t dl_dst[ETH_ADDR_LEN]; /* Ethernet destination address. */ + uint8_t dl_src[ETH_ADDR_LEN]; /* Ethernet source address. */ ovs_be16 dl_type; /* Ethernet frame type. */ ovs_be16 vlan_tci; /* If 802.1Q, TCI | VLAN_CFI; otherwise 0. */ - ovs_be32 mpls_lse[FLOW_MAX_MPLS_LABELS]; /* MPLS label stack entry. */ - - /* L3 */ + ovs_be32 mpls_lse[ROUND_UP(FLOW_MAX_MPLS_LABELS, 2)]; /* MPLS label stack + (with padding). */ + /* L3 (64-bit aligned) */ + ovs_be32 nw_src; /* IPv4 source address. */ + ovs_be32 nw_dst; /* IPv4 destination address. */ struct in6_addr ipv6_src; /* IPv6 source address. */ struct in6_addr ipv6_dst; /* IPv6 destination address. */ ovs_be32 ipv6_label; /* IPv6 flow label. */ - ovs_be32 nw_src; /* IPv4 source address. */ - ovs_be32 nw_dst; /* IPv4 destination address. */ uint8_t nw_frag; /* FLOW_FRAG_* flags. */ uint8_t nw_tos; /* IP ToS (including DSCP and ECN). */ uint8_t nw_ttl; /* IP TTL/Hop Limit. */ uint8_t nw_proto; /* IP protocol or low 8 bits of ARP opcode. */ - uint8_t arp_sha[6]; /* ARP/ND source hardware address. */ - uint8_t arp_tha[6]; /* ARP/ND target hardware address. */ struct in6_addr nd_target; /* IPv6 neighbor discovery (ND) target. */ + uint8_t arp_sha[ETH_ADDR_LEN]; /* ARP/ND source hardware address. */ + uint8_t arp_tha[ETH_ADDR_LEN]; /* ARP/ND target hardware address. */ ovs_be16 tcp_flags; /* TCP flags. With L3 to avoid matching L4. */ - ovs_be16 pad; /* Padding. */ + ovs_be16 pad2; /* Pad to 64 bits. */ - /* L4 */ + /* L4 (64-bit aligned) */ ovs_be16 tp_src; /* TCP/UDP/SCTP source port. */ - ovs_be16 tp_dst; /* TCP/UDP/SCTP destination port. - * Keep last for the BUILD_ASSERT_DECL below */ - uint32_t dp_hash; /* Datapath computed hash value. The exact - computation is opaque to the user space.*/ + ovs_be16 tp_dst; /* TCP/UDP/SCTP destination port. */ + ovs_be32 igmp_group_ip4; /* IGMP group IPv4 address. + * Keep last for BUILD_ASSERT_DECL below. */ }; -BUILD_ASSERT_DECL(sizeof(struct flow) % 4 == 0); +BUILD_ASSERT_DECL(sizeof(struct flow) % sizeof(uint64_t) == 0); + +#define FLOW_U64S (sizeof(struct flow) / sizeof(uint64_t)) -#define FLOW_U32S (sizeof(struct flow) / 4) +/* Some flow fields are mutually exclusive or only appear within the flow + * pipeline. IPv6 headers are bigger than IPv4 and MPLS, and IPv6 ND packets + * are bigger than TCP,UDP and IGMP packets. */ +#define FLOW_MAX_PACKET_U64S (FLOW_U64S \ + /* Unused in datapath */ - FLOW_U64_SIZE(regs) \ + - FLOW_U64_SIZE(metadata) \ + /* L2.5/3 */ - FLOW_U64_SIZE(nw_src) /* incl. nw_dst */ \ + - FLOW_U64_SIZE(mpls_lse) \ + /* L4 */ - FLOW_U64_SIZE(tp_src) \ + ) /* Remember to update FLOW_WC_SEQ when changing 'struct flow'. */ -BUILD_ASSERT_DECL(offsetof(struct flow, dp_hash) + sizeof(uint32_t) - == sizeof(struct flow_tnl) + 172 - && FLOW_WC_SEQ == 26); +BUILD_ASSERT_DECL(offsetof(struct flow, igmp_group_ip4) + sizeof(uint32_t) + == sizeof(struct flow_tnl) + 192 + && FLOW_WC_SEQ == 32); /* Incremental points at which flow classification may be performed in * segments. * This is located here since this is dependent on the structure of the * struct flow defined above: - * Each offset must be on a distinct, successive U32 boundary strictly + * Each offset must be on a distinct, successive U64 boundary strictly * within the struct flow. */ enum { FLOW_SEGMENT_1_ENDS_AT = offsetof(struct flow, dl_dst), - FLOW_SEGMENT_2_ENDS_AT = offsetof(struct flow, ipv6_src), + FLOW_SEGMENT_2_ENDS_AT = offsetof(struct flow, nw_src), FLOW_SEGMENT_3_ENDS_AT = offsetof(struct flow, tp_src), }; -BUILD_ASSERT_DECL(FLOW_SEGMENT_1_ENDS_AT % 4 == 0); -BUILD_ASSERT_DECL(FLOW_SEGMENT_2_ENDS_AT % 4 == 0); -BUILD_ASSERT_DECL(FLOW_SEGMENT_3_ENDS_AT % 4 == 0); +BUILD_ASSERT_DECL(FLOW_SEGMENT_1_ENDS_AT % sizeof(uint64_t) == 0); +BUILD_ASSERT_DECL(FLOW_SEGMENT_2_ENDS_AT % sizeof(uint64_t) == 0); +BUILD_ASSERT_DECL(FLOW_SEGMENT_3_ENDS_AT % sizeof(uint64_t) == 0); BUILD_ASSERT_DECL( 0 < FLOW_SEGMENT_1_ENDS_AT); BUILD_ASSERT_DECL(FLOW_SEGMENT_1_ENDS_AT < FLOW_SEGMENT_2_ENDS_AT); BUILD_ASSERT_DECL(FLOW_SEGMENT_2_ENDS_AT < FLOW_SEGMENT_3_ENDS_AT); BUILD_ASSERT_DECL(FLOW_SEGMENT_3_ENDS_AT < sizeof(struct flow)); -extern const uint8_t flow_segment_u32s[]; - -/* Represents the metadata fields of struct flow. */ -struct flow_metadata { - uint32_t dp_hash; /* Datapath computed hash field. */ - uint32_t recirc_id; /* Recirculation ID. */ - ovs_be64 tun_id; /* Encapsulating tunnel ID. */ - ovs_be32 tun_src; /* Tunnel outer IPv4 src addr */ - ovs_be32 tun_dst; /* Tunnel outer IPv4 dst addr */ - ovs_be64 metadata; /* OpenFlow 1.1+ metadata field. */ - uint32_t regs[FLOW_N_REGS]; /* Registers. */ - uint32_t pkt_mark; /* Packet mark. */ - ofp_port_t in_port; /* OpenFlow port or zero. */ -}; +extern const uint8_t flow_segment_u64s[]; -void flow_extract(struct ofpbuf *, const struct pkt_metadata *md, - struct flow *); +void flow_extract(struct dp_packet *, struct flow *); void flow_zero_wildcards(struct flow *, const struct flow_wildcards *); void flow_unwildcard_tp_ports(const struct flow *, struct flow_wildcards *); -void flow_get_metadata(const struct flow *, struct flow_metadata *); +void flow_get_metadata(const struct flow *, struct match *flow_metadata); char *flow_to_string(const struct flow *); void format_flags(struct ds *ds, const char *(*bit_to_string)(uint32_t), @@ -204,7 +217,20 @@ void flow_set_mpls_tc(struct flow *, int idx, uint8_t tc); void flow_set_mpls_bos(struct flow *, int idx, uint8_t stack); void flow_set_mpls_lse(struct flow *, int idx, ovs_be32 lse); -void flow_compose(struct ofpbuf *, const struct flow *); +void flow_compose(struct dp_packet *, const struct flow *); + +static inline uint64_t +flow_get_xreg(const struct flow *flow, int idx) +{ + return ((uint64_t) flow->regs[idx * 2] << 32) | flow->regs[idx * 2 + 1]; +} + +static inline void +flow_set_xreg(struct flow *flow, int idx, uint64_t value) +{ + flow->regs[idx * 2] = value >> 32; + flow->regs[idx * 2 + 1] = value; +} static inline int flow_compare_3way(const struct flow *a, const struct flow *b) @@ -221,7 +247,8 @@ flow_equal(const struct flow *a, const struct flow *b) static inline size_t flow_hash(const struct flow *flow, uint32_t basis) { - return hash_words((const uint32_t *) flow, sizeof *flow / 4, basis); + return hash_words64((const uint64_t *)flow, + sizeof *flow / sizeof(uint64_t), basis); } static inline uint16_t @@ -271,13 +298,6 @@ hash_odp_port(odp_port_t odp_port) { return hash_int(odp_to_u32(odp_port), 0); } - -uint32_t flow_hash_in_minimask(const struct flow *, const struct minimask *, - uint32_t basis); -uint32_t flow_hash_in_minimask_range(const struct flow *, - const struct minimask *, - uint8_t start, uint8_t end, - uint32_t *basis); /* Wildcards for a flow. * @@ -288,14 +308,25 @@ struct flow_wildcards { struct flow masks; }; +#define WC_MASK_FIELD(WC, FIELD) \ + memset(&(WC)->masks.FIELD, 0xff, sizeof (WC)->masks.FIELD) +#define WC_UNMASK_FIELD(WC, FIELD) \ + memset(&(WC)->masks.FIELD, 0, sizeof (WC)->masks.FIELD) + void flow_wildcards_init_catchall(struct flow_wildcards *); +void flow_wildcards_init_for_packet(struct flow_wildcards *, + const struct flow *); +uint64_t flow_wc_map(const struct flow *); + void flow_wildcards_clear_non_packet_fields(struct flow_wildcards *); bool flow_wildcards_is_catchall(const struct flow_wildcards *); void flow_wildcards_set_reg_mask(struct flow_wildcards *, int idx, uint32_t mask); +void flow_wildcards_set_xreg_mask(struct flow_wildcards *, + int idx, uint64_t mask); void flow_wildcards_and(struct flow_wildcards *dst, const struct flow_wildcards *src1, @@ -305,13 +336,6 @@ void flow_wildcards_or(struct flow_wildcards *dst, const struct flow_wildcards *src2); bool flow_wildcards_has_extra(const struct flow_wildcards *, const struct flow_wildcards *); - -void flow_wildcards_fold_minimask(struct flow_wildcards *, - const struct minimask *); -void flow_wildcards_fold_minimask_range(struct flow_wildcards *, - const struct minimask *, - uint8_t start, uint8_t end); - uint32_t flow_wildcards_hash(const struct flow_wildcards *, uint32_t basis); bool flow_wildcards_equal(const struct flow_wildcards *, const struct flow_wildcards *); @@ -336,8 +360,11 @@ bool flow_equal_except(const struct flow *a, const struct flow *b, /* Compressed flow. */ -#define MINI_N_INLINE (sizeof(void *) == 4 ? 7 : 8) -BUILD_ASSERT_DECL(FLOW_U32S <= 64); +/* Number of 64-bit words present in struct miniflow. */ +#define MINI_N_INLINE 4 + +/* Maximum number of 64-bit words supported. */ +BUILD_ASSERT_DECL(FLOW_U64S <= 63); /* A sparse representation of a "struct flow". * @@ -346,75 +373,207 @@ BUILD_ASSERT_DECL(FLOW_U32S <= 64); * saves time when the goal is to iterate over only the nonzero parts of the * struct. * - * The 'map' member holds one bit for each uint32_t in a "struct flow". Each - * 0-bit indicates that the corresponding uint32_t is zero, each 1-bit that it - * *may* be nonzero. - * - * 'values' points to the start of an array that has one element for each 1-bit - * in 'map'. The least-numbered 1-bit is in values[0], the next 1-bit is in - * values[1], and so on. + * The 'map' member holds one bit for each uint64_t in a "struct flow". Each + * 0-bit indicates that the corresponding uint64_t is zero, each 1-bit that it + * *may* be nonzero (see below how this applies to minimasks). * - * 'values' may point to a few different locations: + * The 'values_inline' boolean member indicates that the values are at + * 'inline_values'. If 'values_inline' is zero, then the values are + * offline at 'offline_values'. In either case, values is an array that has + * one element for each 1-bit in 'map'. The least-numbered 1-bit is in + * the first element of the values array, the next 1-bit is in the next array + * element, and so on. * - * - If 'map' has MINI_N_INLINE or fewer 1-bits, it may point to - * 'inline_values'. One hopes that this is the common case. + * MINI_N_INLINE is the default number of inline words. When a miniflow is + * dynamically allocated the actual amount of inline storage may be different. + * In that case 'inline_values' contains storage at least for the number + * of words indicated by 'map' (one uint64_t for each 1-bit in the map). * - * - If 'map' has more than MINI_N_INLINE 1-bits, it may point to memory - * allocated with malloc(). - * - * - The caller could provide storage on the stack for situations where - * that makes sense. So far that's only proved useful for - * minimask_combine(), but the principle works elsewhere. - * - * Elements in 'values' are allowed to be zero. This is useful for "struct + * Elements in values array are allowed to be zero. This is useful for "struct * minimatch", for which ensuring that the miniflow and minimask members have * same 'map' allows optimization. This allowance applies only to a miniflow * that is not a mask. That is, a minimask may NOT have zero elements in * its 'values'. */ struct miniflow { - uint64_t map; - uint32_t *values; - uint32_t inline_values[MINI_N_INLINE]; + uint64_t map:63; + uint64_t values_inline:1; + union { + uint64_t *offline_values; + uint64_t inline_values[MINI_N_INLINE]; /* Minimum inline size. */ + }; }; +BUILD_ASSERT_DECL(sizeof(struct miniflow) + == sizeof(uint64_t) + MINI_N_INLINE * sizeof(uint64_t)); + +#define MINIFLOW_VALUES_SIZE(COUNT) ((COUNT) * sizeof(uint64_t)) + +static inline uint64_t *miniflow_values(struct miniflow *mf) +{ + return OVS_LIKELY(mf->values_inline) + ? mf->inline_values : mf->offline_values; +} + +static inline const uint64_t *miniflow_get_values(const struct miniflow *mf) +{ + return OVS_LIKELY(mf->values_inline) + ? mf->inline_values : mf->offline_values; +} /* This is useful for initializing a miniflow for a miniflow_extract() call. */ static inline void miniflow_initialize(struct miniflow *mf, - uint32_t buf[FLOW_U32S]) + uint64_t buf[FLOW_U64S]) { mf->map = 0; - mf->values = buf; + mf->values_inline = (buf == (uint64_t *)(mf + 1)); + if (!mf->values_inline) { + mf->offline_values = buf; + } } struct pkt_metadata; /* The 'dst->values' must be initialized with a buffer with space for - * FLOW_U32S. 'dst->map' is ignored on input and set on output to + * FLOW_U64S. 'dst->map' is ignored on input and set on output to * indicate which fields were extracted. */ -void miniflow_extract(struct ofpbuf *packet, const struct pkt_metadata *, - struct miniflow *dst); +void miniflow_extract(struct dp_packet *packet, struct miniflow *dst); void miniflow_init(struct miniflow *, const struct flow *); void miniflow_init_with_minimask(struct miniflow *, const struct flow *, const struct minimask *); void miniflow_clone(struct miniflow *, const struct miniflow *); +void miniflow_clone_inline(struct miniflow *, const struct miniflow *, + size_t n_values); void miniflow_move(struct miniflow *dst, struct miniflow *); void miniflow_destroy(struct miniflow *); void miniflow_expand(const struct miniflow *, struct flow *); -/* These accessors use byte offsets, which are assumed to be compile-time - * constants. */ -static inline uint8_t miniflow_get_u8(const struct miniflow *, - unsigned int ofs); -static inline uint16_t miniflow_get_u16(const struct miniflow *, - unsigned int ofs); -static inline ovs_be16 miniflow_get_be16(const struct miniflow *, - unsigned int ofs); +static inline uint64_t flow_u64_value(const struct flow *flow, size_t index) +{ + return ((uint64_t *)(flow))[index]; +} + +static inline uint64_t *flow_u64_lvalue(struct flow *flow, size_t index) +{ + return &((uint64_t *)(flow))[index]; +} + +static inline bool +flow_get_next_in_map(const struct flow *flow, uint64_t map, uint64_t *value) +{ + if (map) { + *value = flow_u64_value(flow, raw_ctz(map)); + return true; + } + return false; +} + +/* Iterate through all flow u64 values specified by 'MAP'. */ +#define FLOW_FOR_EACH_IN_MAP(VALUE, FLOW, MAP) \ + for (uint64_t map__ = (MAP); \ + flow_get_next_in_map(FLOW, map__, &(VALUE)); \ + map__ = zero_rightmost_1bit(map__)) + +/* Iterate through all struct flow u64 indices specified by 'MAP'. */ +#define MAP_FOR_EACH_INDEX(U64IDX, MAP) \ + for (uint64_t map__ = (MAP); \ + map__ && ((U64IDX) = raw_ctz(map__), true); \ + map__ = zero_rightmost_1bit(map__)) + +#define FLOW_U64_SIZE(FIELD) \ + DIV_ROUND_UP(sizeof(((struct flow *)0)->FIELD), sizeof(uint64_t)) + +#define MINIFLOW_MAP(FIELD) \ + (((UINT64_C(1) << FLOW_U64_SIZE(FIELD)) - 1) \ + << (offsetof(struct flow, FIELD) / sizeof(uint64_t))) + +struct mf_for_each_in_map_aux { + const uint64_t *values; + uint64_t fmap; + uint64_t map; +}; + +static inline bool +mf_get_next_in_map(struct mf_for_each_in_map_aux *aux, uint64_t *value) +{ + if (aux->map) { + uint64_t rm1bit = rightmost_1bit(aux->map); + aux->map -= rm1bit; + + if (aux->fmap & rm1bit) { + /* Advance 'aux->values' to point to the value for 'rm1bit'. */ + uint64_t trash = aux->fmap & (rm1bit - 1); + if (trash) { + aux->fmap -= trash; + aux->values += count_1bits(trash); + } + + /* Retrieve the value for 'rm1bit' then advance past it. */ + aux->fmap -= rm1bit; + *value = *aux->values++; + } else { + *value = 0; + } + return true; + } else { + return false; + } +} + +/* Iterate through all miniflow u64 values specified by 'MAP'. */ +#define MINIFLOW_FOR_EACH_IN_MAP(VALUE, FLOW, MAP) \ + for (struct mf_for_each_in_map_aux aux__ \ + = { miniflow_get_values(FLOW), (FLOW)->map, MAP }; \ + mf_get_next_in_map(&aux__, &(VALUE)); \ + ) + +/* This can be used when it is known that 'u64_idx' is set in 'map'. */ +static inline uint64_t +miniflow_values_get__(const uint64_t *values, uint64_t map, int u64_idx) +{ + return values[count_1bits(map & ((UINT64_C(1) << u64_idx) - 1))]; +} + +/* This can be used when it is known that 'u64_idx' is set in + * the map of 'mf'. */ +static inline uint64_t +miniflow_get__(const struct miniflow *mf, int u64_idx) +{ + return miniflow_values_get__(miniflow_get_values(mf), mf->map, u64_idx); +} + +/* Get the value of 'FIELD' of an up to 8 byte wide integer type 'TYPE' of + * a miniflow. */ +#define MINIFLOW_GET_TYPE(MF, TYPE, OFS) \ + (((MF)->map & (UINT64_C(1) << (OFS) / sizeof(uint64_t))) \ + ? ((OVS_FORCE const TYPE *) \ + (miniflow_get_values(MF) \ + + count_1bits((MF)->map & \ + ((UINT64_C(1) << (OFS) / sizeof(uint64_t)) - 1)))) \ + [(OFS) % sizeof(uint64_t) / sizeof(TYPE)] \ + : 0) \ + +#define MINIFLOW_GET_U8(FLOW, FIELD) \ + MINIFLOW_GET_TYPE(FLOW, uint8_t, offsetof(struct flow, FIELD)) +#define MINIFLOW_GET_U16(FLOW, FIELD) \ + MINIFLOW_GET_TYPE(FLOW, uint16_t, offsetof(struct flow, FIELD)) +#define MINIFLOW_GET_BE16(FLOW, FIELD) \ + MINIFLOW_GET_TYPE(FLOW, ovs_be16, offsetof(struct flow, FIELD)) +#define MINIFLOW_GET_U32(FLOW, FIELD) \ + MINIFLOW_GET_TYPE(FLOW, uint32_t, offsetof(struct flow, FIELD)) +#define MINIFLOW_GET_BE32(FLOW, FIELD) \ + MINIFLOW_GET_TYPE(FLOW, ovs_be32, offsetof(struct flow, FIELD)) +#define MINIFLOW_GET_U64(FLOW, FIELD) \ + MINIFLOW_GET_TYPE(FLOW, uint64_t, offsetof(struct flow, FIELD)) +#define MINIFLOW_GET_BE64(FLOW, FIELD) \ + MINIFLOW_GET_TYPE(FLOW, ovs_be64, offsetof(struct flow, FIELD)) + +static inline uint64_t miniflow_get(const struct miniflow *, + unsigned int u64_ofs); static inline uint32_t miniflow_get_u32(const struct miniflow *, - unsigned int ofs); + unsigned int u32_ofs); static inline ovs_be32 miniflow_get_be32(const struct miniflow *, - unsigned int ofs); - + unsigned int be32_ofs); static inline uint16_t miniflow_get_vid(const struct miniflow *); static inline uint16_t miniflow_get_tcp_flags(const struct miniflow *); static inline ovs_be64 miniflow_get_metadata(const struct miniflow *); @@ -426,12 +585,7 @@ bool miniflow_equal_in_minimask(const struct miniflow *a, bool miniflow_equal_flow_in_minimask(const struct miniflow *a, const struct flow *b, const struct minimask *); -uint32_t miniflow_hash(const struct miniflow *, uint32_t basis); -uint32_t miniflow_hash_in_minimask(const struct miniflow *, - const struct minimask *, uint32_t basis); -uint64_t miniflow_get_map_in_range(const struct miniflow *miniflow, - uint8_t start, uint8_t end, - unsigned int *offset); +uint32_t miniflow_hash_5tuple(const struct miniflow *flow, uint32_t basis); /* Compressed flow wildcards. */ @@ -451,73 +605,83 @@ void minimask_clone(struct minimask *, const struct minimask *); void minimask_move(struct minimask *dst, struct minimask *src); void minimask_combine(struct minimask *dst, const struct minimask *a, const struct minimask *b, - uint32_t storage[FLOW_U32S]); + uint64_t storage[FLOW_U64S]); void minimask_destroy(struct minimask *); void minimask_expand(const struct minimask *, struct flow_wildcards *); -uint32_t minimask_get(const struct minimask *, unsigned int u32_ofs); +static inline uint32_t minimask_get_u32(const struct minimask *, + unsigned int u32_ofs); +static inline ovs_be32 minimask_get_be32(const struct minimask *, + unsigned int be32_ofs); static inline uint16_t minimask_get_vid_mask(const struct minimask *); static inline ovs_be64 minimask_get_metadata_mask(const struct minimask *); bool minimask_equal(const struct minimask *a, const struct minimask *b); -uint32_t minimask_hash(const struct minimask *, uint32_t basis); - bool minimask_has_extra(const struct minimask *, const struct minimask *); -bool minimask_is_catchall(const struct minimask *); - - -/* 'OFS' is a compile-time constant. */ -#define MINIFLOW_GET_TYPE(MF, TYPE, OFS) \ - (MF->map & UINT64_C(1) << OFS / 4) \ - ? ((OVS_FORCE const TYPE *) \ - (MF->values + count_1bits(MF->map & ((UINT64_C(1) << OFS / 4) - 1)))) \ - [OFS % 4 / sizeof(TYPE)] \ - : 0 -static inline uint8_t -miniflow_get_u8(const struct miniflow *flow, unsigned int ofs) + +/* Returns true if 'mask' matches every packet, false if 'mask' fixes any bits + * or fields. */ +static inline bool +minimask_is_catchall(const struct minimask *mask) { - return MINIFLOW_GET_TYPE(flow, uint8_t, ofs); + /* For every 1-bit in mask's map, the corresponding value is non-zero, + * so the only way the mask can not fix any bits or fields is for the + * map the be zero. */ + return mask->masks.map == 0; } -static inline uint16_t -miniflow_get_u16(const struct miniflow *flow, unsigned int ofs) +/* Returns the uint64_t that would be at byte offset '8 * u64_ofs' if 'flow' + * were expanded into a "struct flow". */ +static inline uint64_t miniflow_get(const struct miniflow *flow, + unsigned int u64_ofs) { - return MINIFLOW_GET_TYPE(flow, uint16_t, ofs); + return flow->map & (UINT64_C(1) << u64_ofs) + ? miniflow_get__(flow, u64_ofs) : 0; } -/* Returns the ovs_be16 that would be at byte offset 'u8_ofs' if 'flow' were - * expanded into a "struct flow". */ -static inline ovs_be16 -miniflow_get_be16(const struct miniflow *flow, unsigned int ofs) +static inline uint32_t miniflow_get_u32(const struct miniflow *flow, + unsigned int u32_ofs) { - return MINIFLOW_GET_TYPE(flow, ovs_be16, ofs); -} + uint64_t value = miniflow_get(flow, u32_ofs / 2); -static inline uint32_t -miniflow_get_u32(const struct miniflow *flow, unsigned int ofs) -{ - return MINIFLOW_GET_TYPE(flow, uint32_t, ofs); +#if WORDS_BIGENDIAN + return (u32_ofs & 1) ? value : value >> 32; +#else + return (u32_ofs & 1) ? value >> 32 : value; +#endif } -static inline ovs_be32 -miniflow_get_be32(const struct miniflow *flow, unsigned int ofs) +static inline ovs_be32 miniflow_get_be32(const struct miniflow *flow, + unsigned int be32_ofs) { - return MINIFLOW_GET_TYPE(flow, ovs_be32, ofs); + return (OVS_FORCE ovs_be32)miniflow_get_u32(flow, be32_ofs); } -#undef MINIFLOW_GET_TYPE - /* Returns the VID within the vlan_tci member of the "struct flow" represented * by 'flow'. */ static inline uint16_t miniflow_get_vid(const struct miniflow *flow) { - ovs_be16 tci = miniflow_get_be16(flow, offsetof(struct flow, vlan_tci)); + ovs_be16 tci = MINIFLOW_GET_BE16(flow, vlan_tci); return vlan_tci_to_vid(tci); } +/* Returns the uint32_t that would be at byte offset '4 * u32_ofs' if 'mask' + * were expanded into a "struct flow_wildcards". */ +static inline uint32_t +minimask_get_u32(const struct minimask *mask, unsigned int u32_ofs) +{ + return miniflow_get_u32(&mask->masks, u32_ofs); +} + +static inline ovs_be32 +minimask_get_be32(const struct minimask *mask, unsigned int be32_ofs) +{ + return (OVS_FORCE ovs_be32)minimask_get_u32(mask, be32_ofs); +} + /* Returns the VID mask within the vlan_tci member of the "struct * flow_wildcards" represented by 'mask'. */ static inline uint16_t @@ -530,19 +694,14 @@ minimask_get_vid_mask(const struct minimask *mask) static inline uint16_t miniflow_get_tcp_flags(const struct miniflow *flow) { - return ntohs(miniflow_get_be16(flow, offsetof(struct flow, tcp_flags))); + return ntohs(MINIFLOW_GET_BE16(flow, tcp_flags)); } /* Returns the value of the OpenFlow 1.1+ "metadata" field in 'flow'. */ static inline ovs_be64 miniflow_get_metadata(const struct miniflow *flow) { - enum { MD_OFS = offsetof(struct flow, metadata) }; - BUILD_ASSERT_DECL(MD_OFS % sizeof(uint32_t) == 0); - ovs_be32 hi = miniflow_get_be32(flow, MD_OFS); - ovs_be32 lo = miniflow_get_be32(flow, MD_OFS + 4); - - return htonll(((uint64_t) ntohl(hi) << 32) | ntohl(lo)); + return MINIFLOW_GET_BE64(flow, metadata); } /* Returns the mask for the OpenFlow 1.1+ "metadata" field in 'mask'. @@ -554,22 +713,32 @@ miniflow_get_metadata(const struct miniflow *flow) static inline ovs_be64 minimask_get_metadata_mask(const struct minimask *mask) { - return miniflow_get_metadata(&mask->masks); + return MINIFLOW_GET_BE64(&mask->masks, metadata); } -static inline struct pkt_metadata -pkt_metadata_from_flow(const struct flow *flow) +/* Perform a bitwise OR of miniflow 'src' flow data with the equivalent + * fields in 'dst', storing the result in 'dst'. */ +static inline void +flow_union_with_miniflow(struct flow *dst, const struct miniflow *src) { - struct pkt_metadata md; + uint64_t *dst_u64 = (uint64_t *) dst; + const uint64_t *p = miniflow_get_values(src); + int idx; - md.recirc_id = flow->recirc_id; - md.dp_hash = flow->dp_hash; - md.tunnel = flow->tunnel; - md.skb_priority = flow->skb_priority; - md.pkt_mark = flow->pkt_mark; - md.in_port = flow->in_port; + MAP_FOR_EACH_INDEX(idx, src->map) { + dst_u64[idx] |= *p++; + } +} - return md; +static inline void +pkt_metadata_from_flow(struct pkt_metadata *md, const struct flow *flow) +{ + md->recirc_id = flow->recirc_id; + md->dp_hash = flow->dp_hash; + md->tunnel = flow->tunnel; + md->skb_priority = flow->skb_priority; + md->pkt_mark = flow->pkt_mark; + md->in_port = flow->in_port; } static inline bool is_ip_any(const struct flow *flow) @@ -589,4 +758,35 @@ static inline bool is_icmpv6(const struct flow *flow) && flow->nw_proto == IPPROTO_ICMPV6); } +static inline bool is_igmp(const struct flow *flow) +{ + return (flow->dl_type == htons(ETH_TYPE_IP) + && flow->nw_proto == IPPROTO_IGMP); +} + +static inline bool is_mld(const struct flow *flow) +{ + return is_icmpv6(flow) + && (flow->tp_src == htons(MLD_QUERY) + || flow->tp_src == htons(MLD_REPORT) + || flow->tp_src == htons(MLD_DONE) + || flow->tp_src == htons(MLD2_REPORT)); +} + +static inline bool is_mld_query(const struct flow *flow) +{ + return is_icmpv6(flow) && flow->tp_src == htons(MLD_QUERY); +} + +static inline bool is_mld_report(const struct flow *flow) +{ + return is_mld(flow) && !is_mld_query(flow); +} + +static inline bool is_stp(const struct flow *flow) +{ + return (eth_addr_equals(flow->dl_dst, eth_addr_stp) + && flow->dl_type == htons(FLOW_DL_TYPE_NONE)); +} + #endif /* flow.h */