Commit 29930e31 authored by Jacky Hu's avatar Jacky Hu Committed by Pablo Neira Ayuso

ipvs: add checksum support for gue encapsulation

Add checksum support for gue encapsulation with the tun_flags parameter,
which could be one of the values below:
IP_VS_TUNNEL_ENCAP_FLAG_NOCSUM
IP_VS_TUNNEL_ENCAP_FLAG_CSUM
IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM
Signed-off-by: default avatarJacky Hu <hengqing.hu@gmail.com>
Signed-off-by: default avatarJulian Anastasov <ja@ssi.bg>
Signed-off-by: default avatarSimon Horman <horms@verge.net.au>
Signed-off-by: default avatarPablo Neira Ayuso <pablo@netfilter.org>
parent 2cf6bffc
...@@ -603,6 +603,7 @@ struct ip_vs_dest_user_kern { ...@@ -603,6 +603,7 @@ struct ip_vs_dest_user_kern {
u16 tun_type; /* tunnel type */ u16 tun_type; /* tunnel type */
__be16 tun_port; /* tunnel port */ __be16 tun_port; /* tunnel port */
u16 tun_flags; /* tunnel flags */
}; };
...@@ -665,6 +666,7 @@ struct ip_vs_dest { ...@@ -665,6 +666,7 @@ struct ip_vs_dest {
atomic_t last_weight; /* server latest weight */ atomic_t last_weight; /* server latest weight */
__u16 tun_type; /* tunnel type */ __u16 tun_type; /* tunnel type */
__be16 tun_port; /* tunnel port */ __be16 tun_port; /* tunnel port */
__u16 tun_flags; /* tunnel flags */
refcount_t refcnt; /* reference counter */ refcount_t refcnt; /* reference counter */
struct ip_vs_stats stats; /* statistics */ struct ip_vs_stats stats; /* statistics */
......
...@@ -131,6 +131,11 @@ enum { ...@@ -131,6 +131,11 @@ enum {
IP_VS_CONN_F_TUNNEL_TYPE_MAX, IP_VS_CONN_F_TUNNEL_TYPE_MAX,
}; };
/* Tunnel encapsulation flags */
#define IP_VS_TUNNEL_ENCAP_FLAG_NOCSUM (0)
#define IP_VS_TUNNEL_ENCAP_FLAG_CSUM (1 << 0)
#define IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM (1 << 1)
/* /*
* The struct ip_vs_service_user and struct ip_vs_dest_user are * The struct ip_vs_service_user and struct ip_vs_dest_user are
* used to set IPVS rules through setsockopt. * used to set IPVS rules through setsockopt.
...@@ -403,6 +408,8 @@ enum { ...@@ -403,6 +408,8 @@ enum {
IPVS_DEST_ATTR_TUN_PORT, /* tunnel port */ IPVS_DEST_ATTR_TUN_PORT, /* tunnel port */
IPVS_DEST_ATTR_TUN_FLAGS, /* tunnel flags */
__IPVS_DEST_ATTR_MAX, __IPVS_DEST_ATTR_MAX,
}; };
......
...@@ -893,6 +893,7 @@ __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, ...@@ -893,6 +893,7 @@ __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest,
/* set the tunnel info */ /* set the tunnel info */
dest->tun_type = udest->tun_type; dest->tun_type = udest->tun_type;
dest->tun_port = udest->tun_port; dest->tun_port = udest->tun_port;
dest->tun_flags = udest->tun_flags;
/* set the IP_VS_CONN_F_NOOUTPUT flag if not masquerading/NAT */ /* set the IP_VS_CONN_F_NOOUTPUT flag if not masquerading/NAT */
if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != IP_VS_CONN_F_MASQ) { if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != IP_VS_CONN_F_MASQ) {
...@@ -2967,6 +2968,7 @@ static const struct nla_policy ip_vs_dest_policy[IPVS_DEST_ATTR_MAX + 1] = { ...@@ -2967,6 +2968,7 @@ static const struct nla_policy ip_vs_dest_policy[IPVS_DEST_ATTR_MAX + 1] = {
[IPVS_DEST_ATTR_ADDR_FAMILY] = { .type = NLA_U16 }, [IPVS_DEST_ATTR_ADDR_FAMILY] = { .type = NLA_U16 },
[IPVS_DEST_ATTR_TUN_TYPE] = { .type = NLA_U8 }, [IPVS_DEST_ATTR_TUN_TYPE] = { .type = NLA_U8 },
[IPVS_DEST_ATTR_TUN_PORT] = { .type = NLA_U16 }, [IPVS_DEST_ATTR_TUN_PORT] = { .type = NLA_U16 },
[IPVS_DEST_ATTR_TUN_FLAGS] = { .type = NLA_U16 },
}; };
static int ip_vs_genl_fill_stats(struct sk_buff *skb, int container_type, static int ip_vs_genl_fill_stats(struct sk_buff *skb, int container_type,
...@@ -3273,6 +3275,8 @@ static int ip_vs_genl_fill_dest(struct sk_buff *skb, struct ip_vs_dest *dest) ...@@ -3273,6 +3275,8 @@ static int ip_vs_genl_fill_dest(struct sk_buff *skb, struct ip_vs_dest *dest)
dest->tun_type) || dest->tun_type) ||
nla_put_be16(skb, IPVS_DEST_ATTR_TUN_PORT, nla_put_be16(skb, IPVS_DEST_ATTR_TUN_PORT,
dest->tun_port) || dest->tun_port) ||
nla_put_u16(skb, IPVS_DEST_ATTR_TUN_FLAGS,
dest->tun_flags) ||
nla_put_u32(skb, IPVS_DEST_ATTR_U_THRESH, dest->u_threshold) || nla_put_u32(skb, IPVS_DEST_ATTR_U_THRESH, dest->u_threshold) ||
nla_put_u32(skb, IPVS_DEST_ATTR_L_THRESH, dest->l_threshold) || nla_put_u32(skb, IPVS_DEST_ATTR_L_THRESH, dest->l_threshold) ||
nla_put_u32(skb, IPVS_DEST_ATTR_ACTIVE_CONNS, nla_put_u32(skb, IPVS_DEST_ATTR_ACTIVE_CONNS,
...@@ -3393,7 +3397,8 @@ static int ip_vs_genl_parse_dest(struct ip_vs_dest_user_kern *udest, ...@@ -3393,7 +3397,8 @@ static int ip_vs_genl_parse_dest(struct ip_vs_dest_user_kern *udest,
/* If a full entry was requested, check for the additional fields */ /* If a full entry was requested, check for the additional fields */
if (full_entry) { if (full_entry) {
struct nlattr *nla_fwd, *nla_weight, *nla_u_thresh, struct nlattr *nla_fwd, *nla_weight, *nla_u_thresh,
*nla_l_thresh, *nla_tun_type, *nla_tun_port; *nla_l_thresh, *nla_tun_type, *nla_tun_port,
*nla_tun_flags;
nla_fwd = attrs[IPVS_DEST_ATTR_FWD_METHOD]; nla_fwd = attrs[IPVS_DEST_ATTR_FWD_METHOD];
nla_weight = attrs[IPVS_DEST_ATTR_WEIGHT]; nla_weight = attrs[IPVS_DEST_ATTR_WEIGHT];
...@@ -3401,6 +3406,7 @@ static int ip_vs_genl_parse_dest(struct ip_vs_dest_user_kern *udest, ...@@ -3401,6 +3406,7 @@ static int ip_vs_genl_parse_dest(struct ip_vs_dest_user_kern *udest,
nla_l_thresh = attrs[IPVS_DEST_ATTR_L_THRESH]; nla_l_thresh = attrs[IPVS_DEST_ATTR_L_THRESH];
nla_tun_type = attrs[IPVS_DEST_ATTR_TUN_TYPE]; nla_tun_type = attrs[IPVS_DEST_ATTR_TUN_TYPE];
nla_tun_port = attrs[IPVS_DEST_ATTR_TUN_PORT]; nla_tun_port = attrs[IPVS_DEST_ATTR_TUN_PORT];
nla_tun_flags = attrs[IPVS_DEST_ATTR_TUN_FLAGS];
if (!(nla_fwd && nla_weight && nla_u_thresh && nla_l_thresh)) if (!(nla_fwd && nla_weight && nla_u_thresh && nla_l_thresh))
return -EINVAL; return -EINVAL;
...@@ -3416,6 +3422,9 @@ static int ip_vs_genl_parse_dest(struct ip_vs_dest_user_kern *udest, ...@@ -3416,6 +3422,9 @@ static int ip_vs_genl_parse_dest(struct ip_vs_dest_user_kern *udest,
if (nla_tun_port) if (nla_tun_port)
udest->tun_port = nla_get_be16(nla_tun_port); udest->tun_port = nla_get_be16(nla_tun_port);
if (nla_tun_flags)
udest->tun_flags = nla_get_u16(nla_tun_flags);
} }
return 0; return 0;
......
...@@ -40,6 +40,7 @@ ...@@ -40,6 +40,7 @@
#include <net/ipv6.h> #include <net/ipv6.h>
#include <net/ip6_route.h> #include <net/ip6_route.h>
#include <net/ip_tunnels.h> #include <net/ip_tunnels.h>
#include <net/ip6_checksum.h>
#include <net/addrconf.h> #include <net/addrconf.h>
#include <linux/icmpv6.h> #include <linux/icmpv6.h>
#include <linux/netfilter.h> #include <linux/netfilter.h>
...@@ -385,8 +386,13 @@ __ip_vs_get_out_rt(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, ...@@ -385,8 +386,13 @@ __ip_vs_get_out_rt(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb,
mtu = dst_mtu(&rt->dst) - sizeof(struct iphdr); mtu = dst_mtu(&rt->dst) - sizeof(struct iphdr);
if (!dest) if (!dest)
goto err_put; goto err_put;
if (dest->tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (dest->tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
mtu -= sizeof(struct udphdr) + sizeof(struct guehdr); mtu -= sizeof(struct udphdr) + sizeof(struct guehdr);
if ((dest->tun_flags &
IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM) &&
skb->ip_summed == CHECKSUM_PARTIAL)
mtu -= GUE_PLEN_REMCSUM + GUE_LEN_PRIV;
}
if (mtu < 68) { if (mtu < 68) {
IP_VS_DBG_RL("%s(): mtu less than 68\n", __func__); IP_VS_DBG_RL("%s(): mtu less than 68\n", __func__);
goto err_put; goto err_put;
...@@ -540,8 +546,13 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, ...@@ -540,8 +546,13 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb,
mtu = dst_mtu(&rt->dst) - sizeof(struct ipv6hdr); mtu = dst_mtu(&rt->dst) - sizeof(struct ipv6hdr);
if (!dest) if (!dest)
goto err_put; goto err_put;
if (dest->tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (dest->tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
mtu -= sizeof(struct udphdr) + sizeof(struct guehdr); mtu -= sizeof(struct udphdr) + sizeof(struct guehdr);
if ((dest->tun_flags &
IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM) &&
skb->ip_summed == CHECKSUM_PARTIAL)
mtu -= GUE_PLEN_REMCSUM + GUE_LEN_PRIV;
}
if (mtu < IPV6_MIN_MTU) { if (mtu < IPV6_MIN_MTU) {
IP_VS_DBG_RL("%s(): mtu less than %d\n", __func__, IP_VS_DBG_RL("%s(): mtu less than %d\n", __func__,
IPV6_MIN_MTU); IPV6_MIN_MTU);
...@@ -1006,17 +1017,56 @@ ipvs_gue_encap(struct net *net, struct sk_buff *skb, ...@@ -1006,17 +1017,56 @@ ipvs_gue_encap(struct net *net, struct sk_buff *skb,
__be16 sport = udp_flow_src_port(net, skb, 0, 0, false); __be16 sport = udp_flow_src_port(net, skb, 0, 0, false);
struct udphdr *udph; /* Our new UDP header */ struct udphdr *udph; /* Our new UDP header */
struct guehdr *gueh; /* Our new GUE header */ struct guehdr *gueh; /* Our new GUE header */
size_t hdrlen, optlen = 0;
void *data;
bool need_priv = false;
if ((cp->dest->tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM) &&
skb->ip_summed == CHECKSUM_PARTIAL) {
optlen += GUE_PLEN_REMCSUM + GUE_LEN_PRIV;
need_priv = true;
}
skb_push(skb, sizeof(struct guehdr)); hdrlen = sizeof(struct guehdr) + optlen;
skb_push(skb, hdrlen);
gueh = (struct guehdr *)skb->data; gueh = (struct guehdr *)skb->data;
gueh->control = 0; gueh->control = 0;
gueh->version = 0; gueh->version = 0;
gueh->hlen = 0; gueh->hlen = optlen >> 2;
gueh->flags = 0; gueh->flags = 0;
gueh->proto_ctype = *next_protocol; gueh->proto_ctype = *next_protocol;
data = &gueh[1];
if (need_priv) {
__be32 *flags = data;
u16 csum_start = skb_checksum_start_offset(skb);
__be16 *pd;
gueh->flags |= GUE_FLAG_PRIV;
*flags = 0;
data += GUE_LEN_PRIV;
if (csum_start < hdrlen)
return -EINVAL;
csum_start -= hdrlen;
pd = data;
pd[0] = htons(csum_start);
pd[1] = htons(csum_start + skb->csum_offset);
if (!skb_is_gso(skb)) {
skb->ip_summed = CHECKSUM_NONE;
skb->encapsulation = 0;
}
*flags |= GUE_PFLAG_REMCSUM;
data += GUE_PLEN_REMCSUM;
}
skb_push(skb, sizeof(struct udphdr)); skb_push(skb, sizeof(struct udphdr));
skb_reset_transport_header(skb); skb_reset_transport_header(skb);
...@@ -1070,6 +1120,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1070,6 +1120,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
unsigned int max_headroom; /* The extra header space needed */ unsigned int max_headroom; /* The extra header space needed */
int ret, local; int ret, local;
int tun_type, gso_type; int tun_type, gso_type;
int tun_flags;
EnterFunction(10); EnterFunction(10);
...@@ -1092,9 +1143,19 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1092,9 +1143,19 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct iphdr); max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct iphdr);
tun_type = cp->dest->tun_type; tun_type = cp->dest->tun_type;
tun_flags = cp->dest->tun_flags;
if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
max_headroom += sizeof(struct udphdr) + sizeof(struct guehdr); size_t gue_hdrlen, gue_optlen = 0;
if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM) &&
skb->ip_summed == CHECKSUM_PARTIAL) {
gue_optlen += GUE_PLEN_REMCSUM + GUE_LEN_PRIV;
}
gue_hdrlen = sizeof(struct guehdr) + gue_optlen;
max_headroom += sizeof(struct udphdr) + gue_hdrlen;
}
/* We only care about the df field if sysctl_pmtu_disc(ipvs) is set */ /* We only care about the df field if sysctl_pmtu_disc(ipvs) is set */
dfp = sysctl_pmtu_disc(ipvs) ? &df : NULL; dfp = sysctl_pmtu_disc(ipvs) ? &df : NULL;
...@@ -1105,8 +1166,17 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1105,8 +1166,17 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
goto tx_error; goto tx_error;
gso_type = __tun_gso_type_mask(AF_INET, cp->af); gso_type = __tun_gso_type_mask(AF_INET, cp->af);
if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
gso_type |= SKB_GSO_UDP_TUNNEL; if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) ||
(tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM))
gso_type |= SKB_GSO_UDP_TUNNEL_CSUM;
else
gso_type |= SKB_GSO_UDP_TUNNEL;
if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM) &&
skb->ip_summed == CHECKSUM_PARTIAL) {
gso_type |= SKB_GSO_TUNNEL_REMCSUM;
}
}
if (iptunnel_handle_offloads(skb, gso_type)) if (iptunnel_handle_offloads(skb, gso_type))
goto tx_error; goto tx_error;
...@@ -1115,8 +1185,19 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1115,8 +1185,19 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
skb_set_inner_ipproto(skb, next_protocol); skb_set_inner_ipproto(skb, next_protocol);
if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
ipvs_gue_encap(net, skb, cp, &next_protocol); bool check = false;
if (ipvs_gue_encap(net, skb, cp, &next_protocol))
goto tx_error;
if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) ||
(tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM))
check = true;
udp_set_csum(!check, skb, saddr, cp->daddr.ip, skb->len);
}
skb_push(skb, sizeof(struct iphdr)); skb_push(skb, sizeof(struct iphdr));
skb_reset_network_header(skb); skb_reset_network_header(skb);
...@@ -1174,6 +1255,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1174,6 +1255,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
unsigned int max_headroom; /* The extra header space needed */ unsigned int max_headroom; /* The extra header space needed */
int ret, local; int ret, local;
int tun_type, gso_type; int tun_type, gso_type;
int tun_flags;
EnterFunction(10); EnterFunction(10);
...@@ -1197,9 +1279,19 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1197,9 +1279,19 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct ipv6hdr); max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct ipv6hdr);
tun_type = cp->dest->tun_type; tun_type = cp->dest->tun_type;
tun_flags = cp->dest->tun_flags;
if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
max_headroom += sizeof(struct udphdr) + sizeof(struct guehdr); size_t gue_hdrlen, gue_optlen = 0;
if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM) &&
skb->ip_summed == CHECKSUM_PARTIAL) {
gue_optlen += GUE_PLEN_REMCSUM + GUE_LEN_PRIV;
}
gue_hdrlen = sizeof(struct guehdr) + gue_optlen;
max_headroom += sizeof(struct udphdr) + gue_hdrlen;
}
skb = ip_vs_prepare_tunneled_skb(skb, cp->af, max_headroom, skb = ip_vs_prepare_tunneled_skb(skb, cp->af, max_headroom,
&next_protocol, &payload_len, &next_protocol, &payload_len,
...@@ -1208,8 +1300,17 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1208,8 +1300,17 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
goto tx_error; goto tx_error;
gso_type = __tun_gso_type_mask(AF_INET6, cp->af); gso_type = __tun_gso_type_mask(AF_INET6, cp->af);
if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
gso_type |= SKB_GSO_UDP_TUNNEL; if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) ||
(tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM))
gso_type |= SKB_GSO_UDP_TUNNEL_CSUM;
else
gso_type |= SKB_GSO_UDP_TUNNEL;
if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM) &&
skb->ip_summed == CHECKSUM_PARTIAL) {
gso_type |= SKB_GSO_TUNNEL_REMCSUM;
}
}
if (iptunnel_handle_offloads(skb, gso_type)) if (iptunnel_handle_offloads(skb, gso_type))
goto tx_error; goto tx_error;
...@@ -1218,8 +1319,18 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ...@@ -1218,8 +1319,18 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
skb_set_inner_ipproto(skb, next_protocol); skb_set_inner_ipproto(skb, next_protocol);
if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
ipvs_gue_encap(net, skb, cp, &next_protocol); bool check = false;
if (ipvs_gue_encap(net, skb, cp, &next_protocol))
goto tx_error;
if ((tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) ||
(tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM))
check = true;
udp6_set_csum(!check, skb, &saddr, &cp->daddr.in6, skb->len);
}
skb_push(skb, sizeof(struct ipv6hdr)); skb_push(skb, sizeof(struct ipv6hdr));
skb_reset_network_header(skb); skb_reset_network_header(skb);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment