diff options
author | David S. Miller <davem@davemloft.net> | 2018-08-13 10:07:23 -0700 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2018-08-13 10:07:23 -0700 |
commit | c1617fb4c5eea2ad38b5ffb937a732dbb137117f (patch) | |
tree | 7331bdc1fd66799533d2339816c688ded60e0475 /samples/bpf/xdp_redirect_cpu_kern.c | |
parent | 961d9735357ec59be3e3e2c37c0ca6494f04461b (diff) | |
parent | 2ce3206b9eb3943de09f3bf4ec9134568420d8b9 (diff) |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Daniel Borkmann says:
====================
pull-request: bpf-next 2018-08-13
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Add driver XDP support for veth. This can be used in conjunction with
redirect of another XDP program e.g. sitting on NIC so the xdp_frame
can be forwarded to the peer veth directly without modification,
from Toshiaki.
2) Add a new BPF map type REUSEPORT_SOCKARRAY and prog type SK_REUSEPORT
in order to provide more control and visibility on where a SO_REUSEPORT
sk should be located, and the latter enables to directly select a sk
from the bpf map. This also enables map-in-map for application migration
use cases, from Martin.
3) Add a new BPF helper bpf_skb_ancestor_cgroup_id() that returns the id
of cgroup v2 that is the ancestor of the cgroup associated with the
skb at the ancestor_level, from Andrey.
4) Implement BPF fs map pretty-print support based on BTF data for regular
hash table and LRU map, from Yonghong.
5) Decouple the ability to attach BTF for a map from the key and value
pretty-printer in BPF fs, and enable further support of BTF for maps for
percpu and LPM trie, from Daniel.
6) Implement a better BPF sample of using XDP's CPU redirect feature for
load balancing SKB processing to remote CPU. The sample implements the
same XDP load balancing as Suricata does which is symmetric hash based
on IP and L4 protocol, from Jesper.
7) Revert adding NULL pointer check with WARN_ON_ONCE() in __xdp_return()'s
critical path as it is ensured that the allocator is present, from Björn.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'samples/bpf/xdp_redirect_cpu_kern.c')
-rw-r--r-- | samples/bpf/xdp_redirect_cpu_kern.c | 103 |
1 files changed, 103 insertions, 0 deletions
diff --git a/samples/bpf/xdp_redirect_cpu_kern.c b/samples/bpf/xdp_redirect_cpu_kern.c index 0cc3d71057f0..a306d1c75622 100644 --- a/samples/bpf/xdp_redirect_cpu_kern.c +++ b/samples/bpf/xdp_redirect_cpu_kern.c @@ -13,6 +13,7 @@ #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "hash_func01.h" #define MAX_CPUS 64 /* WARNING - sync with _user.c */ @@ -461,6 +462,108 @@ int xdp_prognum4_ddos_filter_pktgen(struct xdp_md *ctx) return bpf_redirect_map(&cpu_map, cpu_dest, 0); } +/* Hashing initval */ +#define INITVAL 15485863 + +static __always_inline +u32 get_ipv4_hash_ip_pair(struct xdp_md *ctx, u64 nh_off) +{ + void *data_end = (void *)(long)ctx->data_end; + void *data = (void *)(long)ctx->data; + struct iphdr *iph = data + nh_off; + u32 cpu_hash; + + if (iph + 1 > data_end) + return 0; + + cpu_hash = iph->saddr + iph->daddr; + cpu_hash = SuperFastHash((char *)&cpu_hash, 4, INITVAL + iph->protocol); + + return cpu_hash; +} + +static __always_inline +u32 get_ipv6_hash_ip_pair(struct xdp_md *ctx, u64 nh_off) +{ + void *data_end = (void *)(long)ctx->data_end; + void *data = (void *)(long)ctx->data; + struct ipv6hdr *ip6h = data + nh_off; + u32 cpu_hash; + + if (ip6h + 1 > data_end) + return 0; + + cpu_hash = ip6h->saddr.s6_addr32[0] + ip6h->daddr.s6_addr32[0]; + cpu_hash += ip6h->saddr.s6_addr32[1] + ip6h->daddr.s6_addr32[1]; + cpu_hash += ip6h->saddr.s6_addr32[2] + ip6h->daddr.s6_addr32[2]; + cpu_hash += ip6h->saddr.s6_addr32[3] + ip6h->daddr.s6_addr32[3]; + cpu_hash = SuperFastHash((char *)&cpu_hash, 4, INITVAL + ip6h->nexthdr); + + return cpu_hash; +} + +/* Load-Balance traffic based on hashing IP-addrs + L4-proto. The + * hashing scheme is symmetric, meaning swapping IP src/dest still hit + * same CPU. + */ +SEC("xdp_cpu_map5_lb_hash_ip_pairs") +int xdp_prognum5_lb_hash_ip_pairs(struct xdp_md *ctx) +{ + void *data_end = (void *)(long)ctx->data_end; + void *data = (void *)(long)ctx->data; + struct ethhdr *eth = data; + u8 ip_proto = IPPROTO_UDP; + struct datarec *rec; + u16 eth_proto = 0; + u64 l3_offset = 0; + u32 cpu_dest = 0; + u32 cpu_idx = 0; + u32 *cpu_lookup; + u32 *cpu_max; + u32 cpu_hash; + u32 key = 0; + + /* Count RX packet in map */ + rec = bpf_map_lookup_elem(&rx_cnt, &key); + if (!rec) + return XDP_ABORTED; + rec->processed++; + + cpu_max = bpf_map_lookup_elem(&cpus_count, &key); + if (!cpu_max) + return XDP_ABORTED; + + if (!(parse_eth(eth, data_end, ð_proto, &l3_offset))) + return XDP_PASS; /* Just skip */ + + /* Hash for IPv4 and IPv6 */ + switch (eth_proto) { + case ETH_P_IP: + cpu_hash = get_ipv4_hash_ip_pair(ctx, l3_offset); + break; + case ETH_P_IPV6: + cpu_hash = get_ipv6_hash_ip_pair(ctx, l3_offset); + break; + case ETH_P_ARP: /* ARP packet handled on CPU idx 0 */ + default: + cpu_hash = 0; + } + + /* Choose CPU based on hash */ + cpu_idx = cpu_hash % *cpu_max; + + cpu_lookup = bpf_map_lookup_elem(&cpus_available, &cpu_idx); + if (!cpu_lookup) + return XDP_ABORTED; + cpu_dest = *cpu_lookup; + + if (cpu_dest >= MAX_CPUS) { + rec->issue++; + return XDP_ABORTED; + } + + return bpf_redirect_map(&cpu_map, cpu_dest, 0); +} char _license[] SEC("license") = "GPL"; |