diff options
Diffstat (limited to 'net/ipv4')
| -rw-r--r-- | net/ipv4/Makefile | 2 | ||||
| -rw-r--r-- | net/ipv4/bpf_tcp_ca.c | 23 | ||||
| -rw-r--r-- | net/ipv4/fou_bpf.c | 119 | ||||
| -rw-r--r-- | net/ipv4/fou_core.c | 5 | ||||
| -rw-r--r-- | net/ipv4/ip_tunnel.c | 22 | ||||
| -rw-r--r-- | net/ipv4/ipip.c | 1 | ||||
| -rw-r--r-- | net/ipv4/tcp_cong.c | 66 |
7 files changed, 221 insertions, 17 deletions
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 880277c9fd07..b18ba8ef93ad 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -26,7 +26,7 @@ obj-$(CONFIG_IP_MROUTE) += ipmr.o obj-$(CONFIG_IP_MROUTE_COMMON) += ipmr_base.o obj-$(CONFIG_NET_IPIP) += ipip.o gre-y := gre_demux.o -fou-y := fou_core.o fou_nl.o +fou-y := fou_core.o fou_nl.o fou_bpf.o obj-$(CONFIG_NET_FOU) += fou.o obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o obj-$(CONFIG_NET_IPGRE) += ip_gre.o diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index 13fc0c185cd9..4406d796cc2f 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -72,15 +72,11 @@ static bool bpf_tcp_ca_is_valid_access(int off, int size, static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag) + int off, int size) { const struct btf_type *t; size_t end; - if (atype == BPF_READ) - return btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); - t = btf_type_by_id(reg->btf, reg->btf_id); if (t != tcp_sock_type) { bpf_log(log, "only read is supported\n"); @@ -113,6 +109,9 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, case offsetof(struct tcp_sock, ecn_flags): end = offsetofend(struct tcp_sock, ecn_flags); break; + case offsetof(struct tcp_sock, app_limited): + end = offsetofend(struct tcp_sock, app_limited); + break; default: bpf_log(log, "no write support to tcp_sock at off %d\n", off); return -EACCES; @@ -239,8 +238,6 @@ static int bpf_tcp_ca_init_member(const struct btf_type *t, if (bpf_obj_name_cpy(tcp_ca->name, utcp_ca->name, sizeof(tcp_ca->name)) <= 0) return -EINVAL; - if (tcp_ca_find(utcp_ca->name)) - return -EEXIST; return 1; } @@ -266,13 +263,25 @@ static void bpf_tcp_ca_unreg(void *kdata) tcp_unregister_congestion_control(kdata); } +static int bpf_tcp_ca_update(void *kdata, void *old_kdata) +{ + return tcp_update_congestion_control(kdata, old_kdata); +} + +static int bpf_tcp_ca_validate(void *kdata) +{ + return tcp_validate_congestion_control(kdata); +} + struct bpf_struct_ops bpf_tcp_congestion_ops = { .verifier_ops = &bpf_tcp_ca_verifier_ops, .reg = bpf_tcp_ca_reg, .unreg = bpf_tcp_ca_unreg, + .update = bpf_tcp_ca_update, .check_member = bpf_tcp_ca_check_member, .init_member = bpf_tcp_ca_init_member, .init = bpf_tcp_ca_init, + .validate = bpf_tcp_ca_validate, .name = "tcp_congestion_ops", }; diff --git a/net/ipv4/fou_bpf.c b/net/ipv4/fou_bpf.c new file mode 100644 index 000000000000..3760a14b6b57 --- /dev/null +++ b/net/ipv4/fou_bpf.c @@ -0,0 +1,119 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Unstable Fou Helpers for TC-BPF hook + * + * These are called from SCHED_CLS BPF programs. Note that it is + * allowed to break compatibility for these functions since the interface they + * are exposed through to BPF programs is explicitly unstable. + */ + +#include <linux/bpf.h> +#include <linux/btf_ids.h> + +#include <net/dst_metadata.h> +#include <net/fou.h> + +struct bpf_fou_encap { + __be16 sport; + __be16 dport; +}; + +enum bpf_fou_encap_type { + FOU_BPF_ENCAP_FOU, + FOU_BPF_ENCAP_GUE, +}; + +__diag_push(); +__diag_ignore_all("-Wmissing-prototypes", + "Global functions as their definitions will be in BTF"); + +/* bpf_skb_set_fou_encap - Set FOU encap parameters + * + * This function allows for using GUE or FOU encapsulation together with an + * ipip device in collect-metadata mode. + * + * It is meant to be used in BPF tc-hooks and after a call to the + * bpf_skb_set_tunnel_key helper, responsible for setting IP addresses. + * + * Parameters: + * @skb_ctx Pointer to ctx (__sk_buff) in TC program. Cannot be NULL + * @encap Pointer to a `struct bpf_fou_encap` storing UDP src and + * dst ports. If sport is set to 0 the kernel will auto-assign a + * port. This is similar to using `encap-sport auto`. + * Cannot be NULL + * @type Encapsulation type for the packet. Their definitions are + * specified in `enum bpf_fou_encap_type` + */ +__bpf_kfunc int bpf_skb_set_fou_encap(struct __sk_buff *skb_ctx, + struct bpf_fou_encap *encap, int type) +{ + struct sk_buff *skb = (struct sk_buff *)skb_ctx; + struct ip_tunnel_info *info = skb_tunnel_info(skb); + + if (unlikely(!encap)) + return -EINVAL; + + if (unlikely(!info || !(info->mode & IP_TUNNEL_INFO_TX))) + return -EINVAL; + + switch (type) { + case FOU_BPF_ENCAP_FOU: + info->encap.type = TUNNEL_ENCAP_FOU; + break; + case FOU_BPF_ENCAP_GUE: + info->encap.type = TUNNEL_ENCAP_GUE; + break; + default: + info->encap.type = TUNNEL_ENCAP_NONE; + } + + if (info->key.tun_flags & TUNNEL_CSUM) + info->encap.flags |= TUNNEL_ENCAP_FLAG_CSUM; + + info->encap.sport = encap->sport; + info->encap.dport = encap->dport; + + return 0; +} + +/* bpf_skb_get_fou_encap - Get FOU encap parameters + * + * This function allows for reading encap metadata from a packet received + * on an ipip device in collect-metadata mode. + * + * Parameters: + * @skb_ctx Pointer to ctx (__sk_buff) in TC program. Cannot be NULL + * @encap Pointer to a struct bpf_fou_encap storing UDP source and + * destination port. Cannot be NULL + */ +__bpf_kfunc int bpf_skb_get_fou_encap(struct __sk_buff *skb_ctx, + struct bpf_fou_encap *encap) +{ + struct sk_buff *skb = (struct sk_buff *)skb_ctx; + struct ip_tunnel_info *info = skb_tunnel_info(skb); + + if (unlikely(!info)) + return -EINVAL; + + encap->sport = info->encap.sport; + encap->dport = info->encap.dport; + + return 0; +} + +__diag_pop() + +BTF_SET8_START(fou_kfunc_set) +BTF_ID_FLAGS(func, bpf_skb_set_fou_encap) +BTF_ID_FLAGS(func, bpf_skb_get_fou_encap) +BTF_SET8_END(fou_kfunc_set) + +static const struct btf_kfunc_id_set fou_bpf_kfunc_set = { + .owner = THIS_MODULE, + .set = &fou_kfunc_set, +}; + +int register_fou_bpf(void) +{ + return register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, + &fou_bpf_kfunc_set); +} diff --git a/net/ipv4/fou_core.c b/net/ipv4/fou_core.c index cafec9b4eee0..0c41076e31ed 100644 --- a/net/ipv4/fou_core.c +++ b/net/ipv4/fou_core.c @@ -1236,10 +1236,15 @@ static int __init fou_init(void) if (ret < 0) goto unregister; + ret = register_fou_bpf(); + if (ret < 0) + goto kfunc_failed; + ret = ip_tunnel_encap_add_fou_ops(); if (ret == 0) return 0; +kfunc_failed: genl_unregister_family(&fou_nl_family); unregister: unregister_pernet_device(&fou_net_ops); diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 2541083d49ad..beeae624c412 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -359,6 +359,20 @@ err_dev_set_mtu: return ERR_PTR(err); } +void ip_tunnel_md_udp_encap(struct sk_buff *skb, struct ip_tunnel_info *info) +{ + const struct iphdr *iph = ip_hdr(skb); + const struct udphdr *udph; + + if (iph->protocol != IPPROTO_UDP) + return; + + udph = (struct udphdr *)((__u8 *)iph + (iph->ihl << 2)); + info->encap.sport = udph->source; + info->encap.dport = udph->dest; +} +EXPORT_SYMBOL(ip_tunnel_md_udp_encap); + int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst, bool log_ecn_error) @@ -572,7 +586,11 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, tunnel_id_to_key32(key->tun_id), RT_TOS(tos), dev_net(dev), 0, skb->mark, skb_get_hash(skb), key->flow_flags); - if (tunnel->encap.type != TUNNEL_ENCAP_NONE) + + if (!tunnel_hlen) + tunnel_hlen = ip_encap_hlen(&tun_info->encap); + + if (ip_tunnel_encap(skb, &tun_info->encap, &proto, &fl4) < 0) goto tx_error; use_cache = ip_tunnel_dst_cache_usable(skb, tun_info); @@ -732,7 +750,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, dev_net(dev), tunnel->parms.link, tunnel->fwmark, skb_get_hash(skb), 0); - if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0) + if (ip_tunnel_encap(skb, &tunnel->encap, &protocol, &fl4) < 0) goto tx_error; if (connected && md) { diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index abea77759b7e..27b8f83c6ea2 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -241,6 +241,7 @@ static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto) tun_dst = ip_tun_rx_dst(skb, 0, 0, 0); if (!tun_dst) return 0; + ip_tunnel_md_udp_encap(skb, &tun_dst->u.tun_info); } skb_reset_mac_header(skb); diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index db8b4b488c31..1b34050a7538 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -75,14 +75,8 @@ struct tcp_congestion_ops *tcp_ca_find_key(u32 key) return NULL; } -/* - * Attach new congestion control algorithm to the list - * of available options. - */ -int tcp_register_congestion_control(struct tcp_congestion_ops *ca) +int tcp_validate_congestion_control(struct tcp_congestion_ops *ca) { - int ret = 0; - /* all algorithms must implement these */ if (!ca->ssthresh || !ca->undo_cwnd || !(ca->cong_avoid || ca->cong_control)) { @@ -90,6 +84,20 @@ int tcp_register_congestion_control(struct tcp_congestion_ops *ca) return -EINVAL; } + return 0; +} + +/* Attach new congestion control algorithm to the list + * of available options. + */ +int tcp_register_congestion_control(struct tcp_congestion_ops *ca) +{ + int ret; + + ret = tcp_validate_congestion_control(ca); + if (ret) + return ret; + ca->key = jhash(ca->name, sizeof(ca->name), strlen(ca->name)); spin_lock(&tcp_cong_list_lock); @@ -130,6 +138,50 @@ void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca) } EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control); +/* Replace a registered old ca with a new one. + * + * The new ca must have the same name as the old one, that has been + * registered. + */ +int tcp_update_congestion_control(struct tcp_congestion_ops *ca, struct tcp_congestion_ops *old_ca) +{ + struct tcp_congestion_ops *existing; + int ret; + + ret = tcp_validate_congestion_control(ca); + if (ret) + return ret; + + ca->key = jhash(ca->name, sizeof(ca->name), strlen(ca->name)); + + spin_lock(&tcp_cong_list_lock); + existing = tcp_ca_find_key(old_ca->key); + if (ca->key == TCP_CA_UNSPEC || !existing || strcmp(existing->name, ca->name)) { + pr_notice("%s not registered or non-unique key\n", + ca->name); + ret = -EINVAL; + } else if (existing != old_ca) { + pr_notice("invalid old congestion control algorithm to replace\n"); + ret = -EINVAL; + } else { + /* Add the new one before removing the old one to keep + * one implementation available all the time. + */ + list_add_tail_rcu(&ca->list, &tcp_cong_list); + list_del_rcu(&existing->list); + pr_debug("%s updated\n", ca->name); + } + spin_unlock(&tcp_cong_list_lock); + + /* Wait for outstanding readers to complete before the + * module or struct_ops gets removed entirely. + */ + if (!ret) + synchronize_rcu(); + + return ret; +} + u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca) { const struct tcp_congestion_ops *ca; |