From f52f11ec8ad49f697e5158ff87c39b44dab45f51 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 6 Nov 2019 17:01:03 +0800 Subject: [PATCH 1/5] lwtunnel: add options process for arp request Without options copied to the dst tun_info in iptunnel_metadata_reply() called by arp_process for handling arp_request, the generated arp_reply packet may be dropped or sent out with wrong options for some tunnels like erspan and vxlan, and the traffic will break. Fixes: 63d008a4e9ee ("ipv4: send arp replies to the correct tunnel") Signed-off-by: Xin Long Signed-off-by: David S. Miller --- net/ipv4/ip_tunnel_core.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 1452a97914a0..10f08481b003 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -126,15 +126,14 @@ struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, if (!md || md->type != METADATA_IP_TUNNEL || md->u.tun_info.mode & IP_TUNNEL_INFO_TX) - return NULL; - res = metadata_dst_alloc(0, METADATA_IP_TUNNEL, flags); + src = &md->u.tun_info; + res = metadata_dst_alloc(src->options_len, METADATA_IP_TUNNEL, flags); if (!res) return NULL; dst = &res->u.tun_info; - src = &md->u.tun_info; dst->key.tun_id = src->key.tun_id; if (src->mode & IP_TUNNEL_INFO_IPV6) memcpy(&dst->key.u.ipv6.dst, &src->key.u.ipv6.src, @@ -143,6 +142,8 @@ struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, dst->key.u.ipv4.dst = src->key.u.ipv4.src; dst->key.tun_flags = src->key.tun_flags; dst->mode = src->mode | IP_TUNNEL_INFO_TX; + ip_tunnel_info_opts_set(dst, ip_tunnel_info_opts(src), + src->options_len, 0); return res; } From 0eb8eb2f96851e49d54c301aa3b0672706e6b171 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 6 Nov 2019 17:01:04 +0800 Subject: [PATCH 2/5] lwtunnel: add options process for cmp_encap When comparing two tun_info, dst_cache member should have been skipped, as dst_cache is a per cpu pointer and they are always different values even in two tun_info with the same keys. So this patch is to skip dst_cache member and compare the key, mode and options_len only. For the future opts setting support, also to compare options. Fixes: 2d79849903e0 ("lwtunnel: ip tunnel: fix multiple routes with different encap") Signed-off-by: Xin Long Signed-off-by: David S. Miller --- net/ipv4/ip_tunnel_core.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 10f08481b003..c0b5bad8e12a 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -315,8 +315,14 @@ static int ip_tun_encap_nlsize(struct lwtunnel_state *lwtstate) static int ip_tun_cmp_encap(struct lwtunnel_state *a, struct lwtunnel_state *b) { - return memcmp(lwt_tun_info(a), lwt_tun_info(b), - sizeof(struct ip_tunnel_info)); + struct ip_tunnel_info *info_a = lwt_tun_info(a); + struct ip_tunnel_info *info_b = lwt_tun_info(b); + + return memcmp(info_a, info_b, sizeof(info_a->key)) || + info_a->mode != info_b->mode || + info_a->options_len != info_b->options_len || + memcmp(ip_tunnel_info_opts(info_a), + ip_tunnel_info_opts(info_b), info_a->options_len); } static const struct lwtunnel_encap_ops ip_tun_lwt_ops = { From 4ece477870774698e6e73d5821a3dd1605ca123b Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 6 Nov 2019 17:01:05 +0800 Subject: [PATCH 3/5] lwtunnel: add options setting and dumping for geneve To add options setting and dumping, .build_state(), .fill_encap() and .get_encap_size() in ip_tun_lwt_ops needs to be extended: ip_tun_build_state(): ip_tun_parse_opts(): ip_tun_parse_opts_geneve() ip_tun_fill_encap_info(): ip_tun_fill_encap_opts(): ip_tun_fill_encap_opts_geneve() ip_tun_encap_nlsize() ip_tun_opts_nlsize(): if (tun_flags & TUNNEL_GENEVE_OPT) ip_tun_parse_opts(), ip_tun_fill_encap_opts() and ip_tun_opts_nlsize() processes LWTUNNEL_IP_OPTS. ip_tun_parse_opts_geneve(), ip_tun_fill_encap_opts_geneve() and if (tun_flags & TUNNEL_GENEVE_OPT) processes LWTUNNEL_IP_OPTS_GENEVE. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/uapi/linux/lwtunnel.h | 20 ++++ net/ipv4/ip_tunnel_core.c | 212 +++++++++++++++++++++++++++++++--- 2 files changed, 216 insertions(+), 16 deletions(-) diff --git a/include/uapi/linux/lwtunnel.h b/include/uapi/linux/lwtunnel.h index de696ca12f2c..b595ab219036 100644 --- a/include/uapi/linux/lwtunnel.h +++ b/include/uapi/linux/lwtunnel.h @@ -27,6 +27,7 @@ enum lwtunnel_ip_t { LWTUNNEL_IP_TOS, LWTUNNEL_IP_FLAGS, LWTUNNEL_IP_PAD, + LWTUNNEL_IP_OPTS, __LWTUNNEL_IP_MAX, }; @@ -41,11 +42,30 @@ enum lwtunnel_ip6_t { LWTUNNEL_IP6_TC, LWTUNNEL_IP6_FLAGS, LWTUNNEL_IP6_PAD, + LWTUNNEL_IP6_OPTS, __LWTUNNEL_IP6_MAX, }; #define LWTUNNEL_IP6_MAX (__LWTUNNEL_IP6_MAX - 1) +enum { + LWTUNNEL_IP_OPTS_UNSPEC, + LWTUNNEL_IP_OPTS_GENEVE, + __LWTUNNEL_IP_OPTS_MAX, +}; + +#define LWTUNNEL_IP_OPTS_MAX (__LWTUNNEL_IP_OPTS_MAX - 1) + +enum { + LWTUNNEL_IP_OPT_GENEVE_UNSPEC, + LWTUNNEL_IP_OPT_GENEVE_CLASS, + LWTUNNEL_IP_OPT_GENEVE_TYPE, + LWTUNNEL_IP_OPT_GENEVE_DATA, + __LWTUNNEL_IP_OPT_GENEVE_MAX, +}; + +#define LWTUNNEL_IP_OPT_GENEVE_MAX (__LWTUNNEL_IP_OPT_GENEVE_MAX - 1) + enum { LWT_BPF_PROG_UNSPEC, LWT_BPF_PROG_FD, diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index c0b5bad8e12a..1ec9d9419c34 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -34,6 +34,7 @@ #include #include #include +#include const struct ip_tunnel_encap_ops __rcu * iptun_encaps[MAX_IPTUN_ENCAP_OPS] __read_mostly; @@ -218,24 +219,112 @@ static const struct nla_policy ip_tun_policy[LWTUNNEL_IP_MAX + 1] = { [LWTUNNEL_IP_TTL] = { .type = NLA_U8 }, [LWTUNNEL_IP_TOS] = { .type = NLA_U8 }, [LWTUNNEL_IP_FLAGS] = { .type = NLA_U16 }, + [LWTUNNEL_IP_OPTS] = { .type = NLA_NESTED }, }; +static const struct nla_policy ip_opts_policy[LWTUNNEL_IP_OPTS_MAX + 1] = { + [LWTUNNEL_IP_OPTS_GENEVE] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy +geneve_opt_policy[LWTUNNEL_IP_OPT_GENEVE_MAX + 1] = { + [LWTUNNEL_IP_OPT_GENEVE_CLASS] = { .type = NLA_U16 }, + [LWTUNNEL_IP_OPT_GENEVE_TYPE] = { .type = NLA_U8 }, + [LWTUNNEL_IP_OPT_GENEVE_DATA] = { .type = NLA_BINARY, .len = 128 }, +}; + +static int ip_tun_parse_opts_geneve(struct nlattr *attr, + struct ip_tunnel_info *info, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[LWTUNNEL_IP_OPT_GENEVE_MAX + 1]; + int data_len, err; + + err = nla_parse_nested_deprecated(tb, LWTUNNEL_IP_OPT_GENEVE_MAX, + attr, geneve_opt_policy, extack); + if (err) + return err; + + if (!tb[LWTUNNEL_IP_OPT_GENEVE_CLASS] || + !tb[LWTUNNEL_IP_OPT_GENEVE_TYPE] || + !tb[LWTUNNEL_IP_OPT_GENEVE_DATA]) + return -EINVAL; + + attr = tb[LWTUNNEL_IP_OPT_GENEVE_DATA]; + data_len = nla_len(attr); + if (data_len % 4) + return -EINVAL; + + if (info) { + struct geneve_opt *opt = ip_tunnel_info_opts(info); + + memcpy(opt->opt_data, nla_data(attr), data_len); + opt->length = data_len / 4; + attr = tb[LWTUNNEL_IP_OPT_GENEVE_CLASS]; + opt->opt_class = nla_get_be16(attr); + attr = tb[LWTUNNEL_IP_OPT_GENEVE_TYPE]; + opt->type = nla_get_u8(attr); + info->key.tun_flags |= TUNNEL_GENEVE_OPT; + } + + return sizeof(struct geneve_opt) + data_len; +} + +static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[LWTUNNEL_IP_OPTS_MAX + 1]; + int err; + + if (!attr) + return 0; + + err = nla_parse_nested_deprecated(tb, LWTUNNEL_IP_OPTS_MAX, attr, + ip_opts_policy, extack); + if (err) + return err; + + if (tb[LWTUNNEL_IP_OPTS_GENEVE]) + err = ip_tun_parse_opts_geneve(tb[LWTUNNEL_IP_OPTS_GENEVE], + info, extack); + else + err = -EINVAL; + + return err; +} + +static int ip_tun_get_optlen(struct nlattr *attr, + struct netlink_ext_ack *extack) +{ + return ip_tun_parse_opts(attr, NULL, extack); +} + +static int ip_tun_set_opts(struct nlattr *attr, struct ip_tunnel_info *info, + struct netlink_ext_ack *extack) +{ + return ip_tun_parse_opts(attr, info, extack); +} + static int ip_tun_build_state(struct nlattr *attr, unsigned int family, const void *cfg, struct lwtunnel_state **ts, struct netlink_ext_ack *extack) { - struct ip_tunnel_info *tun_info; - struct lwtunnel_state *new_state; struct nlattr *tb[LWTUNNEL_IP_MAX + 1]; - int err; + struct lwtunnel_state *new_state; + struct ip_tunnel_info *tun_info; + int err, opt_len; err = nla_parse_nested_deprecated(tb, LWTUNNEL_IP_MAX, attr, ip_tun_policy, extack); if (err < 0) return err; - new_state = lwtunnel_state_alloc(sizeof(*tun_info)); + opt_len = ip_tun_get_optlen(tb[LWTUNNEL_IP_OPTS], extack); + if (opt_len < 0) + return opt_len; + + new_state = lwtunnel_state_alloc(sizeof(*tun_info) + opt_len); if (!new_state) return -ENOMEM; @@ -243,6 +332,12 @@ static int ip_tun_build_state(struct nlattr *attr, tun_info = lwt_tun_info(new_state); + err = ip_tun_set_opts(tb[LWTUNNEL_IP_OPTS], tun_info, extack); + if (err < 0) { + lwtstate_free(new_state); + return err; + } + #ifdef CONFIG_DST_CACHE err = dst_cache_init(&tun_info->dst_cache, GFP_KERNEL); if (err) { @@ -267,10 +362,10 @@ static int ip_tun_build_state(struct nlattr *attr, tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]); if (tb[LWTUNNEL_IP_FLAGS]) - tun_info->key.tun_flags = nla_get_be16(tb[LWTUNNEL_IP_FLAGS]); + tun_info->key.tun_flags |= nla_get_be16(tb[LWTUNNEL_IP_FLAGS]); tun_info->mode = IP_TUNNEL_INFO_TX; - tun_info->options_len = 0; + tun_info->options_len = opt_len; *ts = new_state; @@ -286,6 +381,54 @@ static void ip_tun_destroy_state(struct lwtunnel_state *lwtstate) #endif } +static int ip_tun_fill_encap_opts_geneve(struct sk_buff *skb, + struct ip_tunnel_info *tun_info) +{ + struct geneve_opt *opt; + struct nlattr *nest; + + nest = nla_nest_start_noflag(skb, LWTUNNEL_IP_OPTS_GENEVE); + if (!nest) + return -ENOMEM; + + opt = ip_tunnel_info_opts(tun_info); + if (nla_put_be16(skb, LWTUNNEL_IP_OPT_GENEVE_CLASS, opt->opt_class) || + nla_put_u8(skb, LWTUNNEL_IP_OPT_GENEVE_TYPE, opt->type) || + nla_put(skb, LWTUNNEL_IP_OPT_GENEVE_DATA, opt->length * 4, + opt->opt_data)) { + nla_nest_cancel(skb, nest); + return -ENOMEM; + } + + nla_nest_end(skb, nest); + return 0; +} + +static int ip_tun_fill_encap_opts(struct sk_buff *skb, int type, + struct ip_tunnel_info *tun_info) +{ + struct nlattr *nest; + int err = 0; + + if (!(tun_info->key.tun_flags & TUNNEL_GENEVE_OPT)) + return 0; + + nest = nla_nest_start_noflag(skb, type); + if (!nest) + return -ENOMEM; + + if (tun_info->key.tun_flags & TUNNEL_GENEVE_OPT) + err = ip_tun_fill_encap_opts_geneve(skb, tun_info); + + if (err) { + nla_nest_cancel(skb, nest); + return err; + } + + nla_nest_end(skb, nest); + return 0; +} + static int ip_tun_fill_encap_info(struct sk_buff *skb, struct lwtunnel_state *lwtstate) { @@ -297,12 +440,34 @@ static int ip_tun_fill_encap_info(struct sk_buff *skb, nla_put_in_addr(skb, LWTUNNEL_IP_SRC, tun_info->key.u.ipv4.src) || nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.tos) || nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ttl) || - nla_put_be16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags)) + nla_put_be16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags) || + ip_tun_fill_encap_opts(skb, LWTUNNEL_IP_OPTS, tun_info)) return -ENOMEM; return 0; } +static int ip_tun_opts_nlsize(struct ip_tunnel_info *info) +{ + int opt_len; + + if (!(info->key.tun_flags & TUNNEL_GENEVE_OPT)) + return 0; + + opt_len = nla_total_size(0); /* LWTUNNEL_IP_OPTS */ + if (info->key.tun_flags & TUNNEL_GENEVE_OPT) { + struct geneve_opt *opt = ip_tunnel_info_opts(info); + + opt_len += nla_total_size(0) /* LWTUNNEL_IP_OPTS_GENEVE */ + + nla_total_size(2) /* OPT_GENEVE_CLASS */ + + nla_total_size(1) /* OPT_GENEVE_TYPE */ + + nla_total_size(opt->length * 4); + /* OPT_GENEVE_DATA */ + } + + return opt_len; +} + static int ip_tun_encap_nlsize(struct lwtunnel_state *lwtstate) { return nla_total_size_64bit(8) /* LWTUNNEL_IP_ID */ @@ -310,7 +475,9 @@ static int ip_tun_encap_nlsize(struct lwtunnel_state *lwtstate) + nla_total_size(4) /* LWTUNNEL_IP_SRC */ + nla_total_size(1) /* LWTUNNEL_IP_TOS */ + nla_total_size(1) /* LWTUNNEL_IP_TTL */ - + nla_total_size(2); /* LWTUNNEL_IP_FLAGS */ + + nla_total_size(2) /* LWTUNNEL_IP_FLAGS */ + + ip_tun_opts_nlsize(lwt_tun_info(lwtstate)); + /* LWTUNNEL_IP_OPTS */ } static int ip_tun_cmp_encap(struct lwtunnel_state *a, struct lwtunnel_state *b) @@ -348,17 +515,21 @@ static int ip6_tun_build_state(struct nlattr *attr, struct lwtunnel_state **ts, struct netlink_ext_ack *extack) { - struct ip_tunnel_info *tun_info; - struct lwtunnel_state *new_state; struct nlattr *tb[LWTUNNEL_IP6_MAX + 1]; - int err; + struct lwtunnel_state *new_state; + struct ip_tunnel_info *tun_info; + int err, opt_len; err = nla_parse_nested_deprecated(tb, LWTUNNEL_IP6_MAX, attr, ip6_tun_policy, extack); if (err < 0) return err; - new_state = lwtunnel_state_alloc(sizeof(*tun_info)); + opt_len = ip_tun_get_optlen(tb[LWTUNNEL_IP6_OPTS], extack); + if (opt_len < 0) + return opt_len; + + new_state = lwtunnel_state_alloc(sizeof(*tun_info) + opt_len); if (!new_state) return -ENOMEM; @@ -366,6 +537,12 @@ static int ip6_tun_build_state(struct nlattr *attr, tun_info = lwt_tun_info(new_state); + err = ip_tun_set_opts(tb[LWTUNNEL_IP6_OPTS], tun_info, extack); + if (err < 0) { + lwtstate_free(new_state); + return err; + } + if (tb[LWTUNNEL_IP6_ID]) tun_info->key.tun_id = nla_get_be64(tb[LWTUNNEL_IP6_ID]); @@ -382,10 +559,10 @@ static int ip6_tun_build_state(struct nlattr *attr, tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP6_TC]); if (tb[LWTUNNEL_IP6_FLAGS]) - tun_info->key.tun_flags = nla_get_be16(tb[LWTUNNEL_IP6_FLAGS]); + tun_info->key.tun_flags |= nla_get_be16(tb[LWTUNNEL_IP6_FLAGS]); tun_info->mode = IP_TUNNEL_INFO_TX | IP_TUNNEL_INFO_IPV6; - tun_info->options_len = 0; + tun_info->options_len = opt_len; *ts = new_state; @@ -403,7 +580,8 @@ static int ip6_tun_fill_encap_info(struct sk_buff *skb, nla_put_in6_addr(skb, LWTUNNEL_IP6_SRC, &tun_info->key.u.ipv6.src) || nla_put_u8(skb, LWTUNNEL_IP6_TC, tun_info->key.tos) || nla_put_u8(skb, LWTUNNEL_IP6_HOPLIMIT, tun_info->key.ttl) || - nla_put_be16(skb, LWTUNNEL_IP6_FLAGS, tun_info->key.tun_flags)) + nla_put_be16(skb, LWTUNNEL_IP6_FLAGS, tun_info->key.tun_flags) || + ip_tun_fill_encap_opts(skb, LWTUNNEL_IP6_OPTS, tun_info)) return -ENOMEM; return 0; @@ -416,7 +594,9 @@ static int ip6_tun_encap_nlsize(struct lwtunnel_state *lwtstate) + nla_total_size(16) /* LWTUNNEL_IP6_SRC */ + nla_total_size(1) /* LWTUNNEL_IP6_HOPLIMIT */ + nla_total_size(1) /* LWTUNNEL_IP6_TC */ - + nla_total_size(2); /* LWTUNNEL_IP6_FLAGS */ + + nla_total_size(2) /* LWTUNNEL_IP6_FLAGS */ + + ip_tun_opts_nlsize(lwt_tun_info(lwtstate)); + /* LWTUNNEL_IP6_OPTS */ } static const struct lwtunnel_encap_ops ip6_tun_lwt_ops = { From edf31cbb1502481da181a09148adb33e12599185 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 6 Nov 2019 17:01:06 +0800 Subject: [PATCH 4/5] lwtunnel: add options setting and dumping for vxlan Based on the code framework built on the last patch, to support setting and dumping for vxlan, we only need to add ip_tun_parse_opts_vxlan() for .build_state and ip_tun_fill_encap_opts_vxlan() for .fill_encap and if (tun_flags & TUNNEL_VXLAN_OPT) for .get_encap_size. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/uapi/linux/lwtunnel.h | 9 +++++ net/ipv4/ip_tunnel_core.c | 67 +++++++++++++++++++++++++++++++++-- 2 files changed, 74 insertions(+), 2 deletions(-) diff --git a/include/uapi/linux/lwtunnel.h b/include/uapi/linux/lwtunnel.h index b595ab219036..638b7b108453 100644 --- a/include/uapi/linux/lwtunnel.h +++ b/include/uapi/linux/lwtunnel.h @@ -51,6 +51,7 @@ enum lwtunnel_ip6_t { enum { LWTUNNEL_IP_OPTS_UNSPEC, LWTUNNEL_IP_OPTS_GENEVE, + LWTUNNEL_IP_OPTS_VXLAN, __LWTUNNEL_IP_OPTS_MAX, }; @@ -66,6 +67,14 @@ enum { #define LWTUNNEL_IP_OPT_GENEVE_MAX (__LWTUNNEL_IP_OPT_GENEVE_MAX - 1) +enum { + LWTUNNEL_IP_OPT_VXLAN_UNSPEC, + LWTUNNEL_IP_OPT_VXLAN_GBP, + __LWTUNNEL_IP_OPT_VXLAN_MAX, +}; + +#define LWTUNNEL_IP_OPT_VXLAN_MAX (__LWTUNNEL_IP_OPT_VXLAN_MAX - 1) + enum { LWT_BPF_PROG_UNSPEC, LWT_BPF_PROG_FD, diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 1ec9d9419c34..61be2e0cbb19 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -35,6 +35,7 @@ #include #include #include +#include const struct ip_tunnel_encap_ops __rcu * iptun_encaps[MAX_IPTUN_ENCAP_OPS] __read_mostly; @@ -224,6 +225,7 @@ static const struct nla_policy ip_tun_policy[LWTUNNEL_IP_MAX + 1] = { static const struct nla_policy ip_opts_policy[LWTUNNEL_IP_OPTS_MAX + 1] = { [LWTUNNEL_IP_OPTS_GENEVE] = { .type = NLA_NESTED }, + [LWTUNNEL_IP_OPTS_VXLAN] = { .type = NLA_NESTED }, }; static const struct nla_policy @@ -233,6 +235,11 @@ geneve_opt_policy[LWTUNNEL_IP_OPT_GENEVE_MAX + 1] = { [LWTUNNEL_IP_OPT_GENEVE_DATA] = { .type = NLA_BINARY, .len = 128 }, }; +static const struct nla_policy +vxlan_opt_policy[LWTUNNEL_IP_OPT_VXLAN_MAX + 1] = { + [LWTUNNEL_IP_OPT_VXLAN_GBP] = { .type = NLA_U32 }, +}; + static int ip_tun_parse_opts_geneve(struct nlattr *attr, struct ip_tunnel_info *info, struct netlink_ext_ack *extack) @@ -270,6 +277,32 @@ static int ip_tun_parse_opts_geneve(struct nlattr *attr, return sizeof(struct geneve_opt) + data_len; } +static int ip_tun_parse_opts_vxlan(struct nlattr *attr, + struct ip_tunnel_info *info, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[LWTUNNEL_IP_OPT_VXLAN_MAX + 1]; + int err; + + err = nla_parse_nested_deprecated(tb, LWTUNNEL_IP_OPT_VXLAN_MAX, + attr, vxlan_opt_policy, extack); + if (err) + return err; + + if (!tb[LWTUNNEL_IP_OPT_VXLAN_GBP]) + return -EINVAL; + + if (info) { + struct vxlan_metadata *md = ip_tunnel_info_opts(info); + + attr = tb[LWTUNNEL_IP_OPT_VXLAN_GBP]; + md->gbp = nla_get_u32(attr); + info->key.tun_flags |= TUNNEL_VXLAN_OPT; + } + + return sizeof(struct vxlan_metadata); +} + static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, struct netlink_ext_ack *extack) { @@ -287,6 +320,9 @@ static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, if (tb[LWTUNNEL_IP_OPTS_GENEVE]) err = ip_tun_parse_opts_geneve(tb[LWTUNNEL_IP_OPTS_GENEVE], info, extack); + else if (tb[LWTUNNEL_IP_OPTS_VXLAN]) + err = ip_tun_parse_opts_vxlan(tb[LWTUNNEL_IP_OPTS_VXLAN], + info, extack); else err = -EINVAL; @@ -404,13 +440,34 @@ static int ip_tun_fill_encap_opts_geneve(struct sk_buff *skb, return 0; } +static int ip_tun_fill_encap_opts_vxlan(struct sk_buff *skb, + struct ip_tunnel_info *tun_info) +{ + struct vxlan_metadata *md; + struct nlattr *nest; + + nest = nla_nest_start_noflag(skb, LWTUNNEL_IP_OPTS_VXLAN); + if (!nest) + return -ENOMEM; + + md = ip_tunnel_info_opts(tun_info); + if (nla_put_u32(skb, LWTUNNEL_IP_OPT_VXLAN_GBP, md->gbp)) { + nla_nest_cancel(skb, nest); + return -ENOMEM; + } + + nla_nest_end(skb, nest); + return 0; +} + static int ip_tun_fill_encap_opts(struct sk_buff *skb, int type, struct ip_tunnel_info *tun_info) { struct nlattr *nest; int err = 0; - if (!(tun_info->key.tun_flags & TUNNEL_GENEVE_OPT)) + if (!(tun_info->key.tun_flags & + (TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT))) return 0; nest = nla_nest_start_noflag(skb, type); @@ -419,6 +476,8 @@ static int ip_tun_fill_encap_opts(struct sk_buff *skb, int type, if (tun_info->key.tun_flags & TUNNEL_GENEVE_OPT) err = ip_tun_fill_encap_opts_geneve(skb, tun_info); + else if (tun_info->key.tun_flags & TUNNEL_VXLAN_OPT) + err = ip_tun_fill_encap_opts_vxlan(skb, tun_info); if (err) { nla_nest_cancel(skb, nest); @@ -451,7 +510,8 @@ static int ip_tun_opts_nlsize(struct ip_tunnel_info *info) { int opt_len; - if (!(info->key.tun_flags & TUNNEL_GENEVE_OPT)) + if (!(info->key.tun_flags & + (TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT))) return 0; opt_len = nla_total_size(0); /* LWTUNNEL_IP_OPTS */ @@ -463,6 +523,9 @@ static int ip_tun_opts_nlsize(struct ip_tunnel_info *info) + nla_total_size(1) /* OPT_GENEVE_TYPE */ + nla_total_size(opt->length * 4); /* OPT_GENEVE_DATA */ + } else if (info->key.tun_flags & TUNNEL_VXLAN_OPT) { + opt_len += nla_total_size(0) /* LWTUNNEL_IP_OPTS_VXLAN */ + + nla_total_size(4); /* OPT_VXLAN_GBP */ } return opt_len; From b0a21810bd5e1f92e3379899cc8ca9fe144ee8b3 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 6 Nov 2019 17:01:07 +0800 Subject: [PATCH 5/5] lwtunnel: add options setting and dumping for erspan Based on the code framework built on the last patch, to support setting and dumping for vxlan, we only need to add ip_tun_parse_opts_erspan() for .build_state and ip_tun_fill_encap_opts_erspan() for .fill_encap and if (tun_flags & TUNNEL_ERSPAN_OPT) for .get_encap_size. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/uapi/linux/lwtunnel.h | 12 +++++ net/ipv4/ip_tunnel_core.c | 94 ++++++++++++++++++++++++++++++++++- 2 files changed, 104 insertions(+), 2 deletions(-) diff --git a/include/uapi/linux/lwtunnel.h b/include/uapi/linux/lwtunnel.h index 638b7b108453..f6035f737193 100644 --- a/include/uapi/linux/lwtunnel.h +++ b/include/uapi/linux/lwtunnel.h @@ -52,6 +52,7 @@ enum { LWTUNNEL_IP_OPTS_UNSPEC, LWTUNNEL_IP_OPTS_GENEVE, LWTUNNEL_IP_OPTS_VXLAN, + LWTUNNEL_IP_OPTS_ERSPAN, __LWTUNNEL_IP_OPTS_MAX, }; @@ -75,6 +76,17 @@ enum { #define LWTUNNEL_IP_OPT_VXLAN_MAX (__LWTUNNEL_IP_OPT_VXLAN_MAX - 1) +enum { + LWTUNNEL_IP_OPT_ERSPAN_UNSPEC, + LWTUNNEL_IP_OPT_ERSPAN_VER, + LWTUNNEL_IP_OPT_ERSPAN_INDEX, + LWTUNNEL_IP_OPT_ERSPAN_DIR, + LWTUNNEL_IP_OPT_ERSPAN_HWID, + __LWTUNNEL_IP_OPT_ERSPAN_MAX, +}; + +#define LWTUNNEL_IP_OPT_ERSPAN_MAX (__LWTUNNEL_IP_OPT_ERSPAN_MAX - 1) + enum { LWT_BPF_PROG_UNSPEC, LWT_BPF_PROG_FD, diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 61be2e0cbb19..d4f84bf9289a 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -36,6 +36,7 @@ #include #include #include +#include const struct ip_tunnel_encap_ops __rcu * iptun_encaps[MAX_IPTUN_ENCAP_OPS] __read_mostly; @@ -226,6 +227,7 @@ static const struct nla_policy ip_tun_policy[LWTUNNEL_IP_MAX + 1] = { static const struct nla_policy ip_opts_policy[LWTUNNEL_IP_OPTS_MAX + 1] = { [LWTUNNEL_IP_OPTS_GENEVE] = { .type = NLA_NESTED }, [LWTUNNEL_IP_OPTS_VXLAN] = { .type = NLA_NESTED }, + [LWTUNNEL_IP_OPTS_ERSPAN] = { .type = NLA_NESTED }, }; static const struct nla_policy @@ -240,6 +242,14 @@ vxlan_opt_policy[LWTUNNEL_IP_OPT_VXLAN_MAX + 1] = { [LWTUNNEL_IP_OPT_VXLAN_GBP] = { .type = NLA_U32 }, }; +static const struct nla_policy +erspan_opt_policy[LWTUNNEL_IP_OPT_ERSPAN_MAX + 1] = { + [LWTUNNEL_IP_OPT_ERSPAN_VER] = { .type = NLA_U8 }, + [LWTUNNEL_IP_OPT_ERSPAN_INDEX] = { .type = NLA_U32 }, + [LWTUNNEL_IP_OPT_ERSPAN_DIR] = { .type = NLA_U8 }, + [LWTUNNEL_IP_OPT_ERSPAN_HWID] = { .type = NLA_U8 }, +}; + static int ip_tun_parse_opts_geneve(struct nlattr *attr, struct ip_tunnel_info *info, struct netlink_ext_ack *extack) @@ -303,6 +313,46 @@ static int ip_tun_parse_opts_vxlan(struct nlattr *attr, return sizeof(struct vxlan_metadata); } +static int ip_tun_parse_opts_erspan(struct nlattr *attr, + struct ip_tunnel_info *info, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[LWTUNNEL_IP_OPT_ERSPAN_MAX + 1]; + int err; + + err = nla_parse_nested_deprecated(tb, LWTUNNEL_IP_OPT_ERSPAN_MAX, + attr, erspan_opt_policy, extack); + if (err) + return err; + + if (!tb[LWTUNNEL_IP_OPT_ERSPAN_VER]) + return -EINVAL; + + if (info) { + struct erspan_metadata *md = ip_tunnel_info_opts(info); + + attr = tb[LWTUNNEL_IP_OPT_ERSPAN_VER]; + md->version = nla_get_u8(attr); + + if (md->version == 1 && tb[LWTUNNEL_IP_OPT_ERSPAN_INDEX]) { + attr = tb[LWTUNNEL_IP_OPT_ERSPAN_INDEX]; + md->u.index = nla_get_be32(attr); + } else if (md->version == 2 && tb[LWTUNNEL_IP_OPT_ERSPAN_DIR] && + tb[LWTUNNEL_IP_OPT_ERSPAN_HWID]) { + attr = tb[LWTUNNEL_IP_OPT_ERSPAN_DIR]; + md->u.md2.dir = nla_get_u8(attr); + attr = tb[LWTUNNEL_IP_OPT_ERSPAN_HWID]; + set_hwid(&md->u.md2, nla_get_u8(attr)); + } else { + return -EINVAL; + } + + info->key.tun_flags |= TUNNEL_ERSPAN_OPT; + } + + return sizeof(struct erspan_metadata); +} + static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, struct netlink_ext_ack *extack) { @@ -323,6 +373,9 @@ static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, else if (tb[LWTUNNEL_IP_OPTS_VXLAN]) err = ip_tun_parse_opts_vxlan(tb[LWTUNNEL_IP_OPTS_VXLAN], info, extack); + else if (tb[LWTUNNEL_IP_OPTS_ERSPAN]) + err = ip_tun_parse_opts_erspan(tb[LWTUNNEL_IP_OPTS_ERSPAN], + info, extack); else err = -EINVAL; @@ -460,6 +513,37 @@ static int ip_tun_fill_encap_opts_vxlan(struct sk_buff *skb, return 0; } +static int ip_tun_fill_encap_opts_erspan(struct sk_buff *skb, + struct ip_tunnel_info *tun_info) +{ + struct erspan_metadata *md; + struct nlattr *nest; + + nest = nla_nest_start_noflag(skb, LWTUNNEL_IP_OPTS_ERSPAN); + if (!nest) + return -ENOMEM; + + md = ip_tunnel_info_opts(tun_info); + if (nla_put_u32(skb, LWTUNNEL_IP_OPT_ERSPAN_VER, md->version)) + goto err; + + if (md->version == 1 && + nla_put_be32(skb, LWTUNNEL_IP_OPT_ERSPAN_INDEX, md->u.index)) + goto err; + + if (md->version == 2 && + (nla_put_u8(skb, LWTUNNEL_IP_OPT_ERSPAN_DIR, md->u.md2.dir) || + nla_put_u8(skb, LWTUNNEL_IP_OPT_ERSPAN_HWID, + get_hwid(&md->u.md2)))) + goto err; + + nla_nest_end(skb, nest); + return 0; +err: + nla_nest_cancel(skb, nest); + return -ENOMEM; +} + static int ip_tun_fill_encap_opts(struct sk_buff *skb, int type, struct ip_tunnel_info *tun_info) { @@ -467,7 +551,7 @@ static int ip_tun_fill_encap_opts(struct sk_buff *skb, int type, int err = 0; if (!(tun_info->key.tun_flags & - (TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT))) + (TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT | TUNNEL_ERSPAN_OPT))) return 0; nest = nla_nest_start_noflag(skb, type); @@ -478,6 +562,8 @@ static int ip_tun_fill_encap_opts(struct sk_buff *skb, int type, err = ip_tun_fill_encap_opts_geneve(skb, tun_info); else if (tun_info->key.tun_flags & TUNNEL_VXLAN_OPT) err = ip_tun_fill_encap_opts_vxlan(skb, tun_info); + else if (tun_info->key.tun_flags & TUNNEL_ERSPAN_OPT) + err = ip_tun_fill_encap_opts_erspan(skb, tun_info); if (err) { nla_nest_cancel(skb, nest); @@ -511,7 +597,7 @@ static int ip_tun_opts_nlsize(struct ip_tunnel_info *info) int opt_len; if (!(info->key.tun_flags & - (TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT))) + (TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT | TUNNEL_ERSPAN_OPT))) return 0; opt_len = nla_total_size(0); /* LWTUNNEL_IP_OPTS */ @@ -526,6 +612,10 @@ static int ip_tun_opts_nlsize(struct ip_tunnel_info *info) } else if (info->key.tun_flags & TUNNEL_VXLAN_OPT) { opt_len += nla_total_size(0) /* LWTUNNEL_IP_OPTS_VXLAN */ + nla_total_size(4); /* OPT_VXLAN_GBP */ + } else if (info->key.tun_flags & TUNNEL_ERSPAN_OPT) { + opt_len += nla_total_size(0) /* LWTUNNEL_IP_OPTS_ERSPAN */ + + nla_total_size(1) /* OPT_ERSPAN_VER */ + + nla_total_size(4); /* OPT_ERSPAN_INDEX/DIR/HWID */ } return opt_len;