[PATCH] ipvsadm: allow tunneling with gue encapsulation

[Jacky Hu <hengqing.hu@xxxxxxxxx>]

Added the following two options:
--gue-port with adding and editing destinations for tunneling servers.
--gue with listing services.

Signed-off-by: Jacky Hu <hengqing.hu@xxxxxxxxx>
---
 ipvsadm.c         | 109 +++++++++++++++++++++++++++++++++++++---------
 libipvs/ip_vs.h   |  16 +++++++
 libipvs/libipvs.c |   3 +-
 3 files changed, 106 insertions(+), 22 deletions(-)

diff --git a/ipvsadm.c b/ipvsadm.c
index 0cb2b68..a3d77d1 100644
--- a/ipvsadm.c
+++ b/ipvsadm.c
@@ -187,7 +187,9 @@ static const char* cmdnames[] = {
 #define OPT_MCAST_PORT         0x02000000
 #define OPT_MCAST_TTL          0x04000000
 #define OPT_SYNC_MAXLEN        0x08000000
-#define NUMBER_OF_OPT          28
+#define OPT_GUE_PORT   0x10000000
+#define OPT_GUE        0x20000000
+#define NUMBER_OF_OPT          30
 
 static const char* optnames[] = {
        "numeric",
@@ -218,6 +220,8 @@ static const char* optnames[] = {
        "mcast-port",
        "mcast-ttl",
        "sync-maxlen",
+       "gue-port",
+       "gue",
 };
 
 /*
@@ -230,21 +234,21 @@ static const char* optnames[] = {
  */
 static const char commands_v_options[NUMBER_OF_CMD][NUMBER_OF_OPT] =
 {
-       /*   -n   -c   svc  -s   -p   -M   -r   fwd  -w   -x   -y   -mc  tot  
dmn  -st  -rt  thr  -pc  srt  sid  -ex  ops  -pe  -b   grp  port ttl  size */
-/*ADD*/     {'x', 'x', '+', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', ' ', ' ', 'x', 'x', 'x', 'x'},
-/*EDIT*/    {'x', 'x', '+', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', ' ', ' ', 'x', 'x', 'x', 'x'},
-/*DEL*/     {'x', 'x', '+', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*FLUSH*/   {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*LIST*/    {' ', '1', '1', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', '1', 
'1', ' ', ' ', ' ', ' ', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*ADDSRV*/  {'x', 'x', '+', 'x', 'x', 'x', '+', ' ', ' ', ' ', ' ', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*DELSRV*/  {'x', 'x', '+', 'x', 'x', 'x', '+', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*EDITSRV*/ {'x', 'x', '+', 'x', 'x', 'x', '+', ' ', ' ', ' ', ' ', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*TIMEOUT*/ {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*STARTD*/  {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', ' ', 'x', 'x', 'x', 'x', ' ', ' ', ' ', ' '},
-/*STOPD*/   {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*RESTORE*/ {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*SAVE*/    {' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
-/*ZERO*/    {'x', 'x', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x'},
+       /*   -n   -c   svc  -s   -p   -M   -r   fwd  -w   -x   -y   -mc  tot  
dmn  -st  -rt  thr  -pc  srt  sid  -ex  ops  -pe  -b   grp  port ttl  size  
gue-port  gue */
+/*ADD*/     {'x', 'x', '+', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*EDIT*/    {'x', 'x', '+', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*DEL*/     {'x', 'x', '+', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*FLUSH*/   {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*LIST*/    {' ', '1', '1', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', '1', 
'1', ' ', ' ', ' ', ' ', ' ', ' ', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
' '},
+/*ADDSRV*/  {'x', 'x', '+', 'x', 'x', 'x', '+', ' ', ' ', ' ', ' ', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', 
'x'},
+/*DELSRV*/  {'x', 'x', '+', 'x', 'x', 'x', '+', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*EDITSRV*/ {'x', 'x', '+', 'x', 'x', 'x', '+', ' ', ' ', ' ', ' ', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', 
'x'},
+/*TIMEOUT*/ {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*STARTD*/  {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', ' ', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', ' ', 'x', 'x', 'x', 'x', ' ', ' ', ' ', ' ', 'x', 
'x'},
+/*STOPD*/   {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*RESTORE*/ {'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*SAVE*/    {' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
+/*ZERO*/    {'x', 'x', ' ', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 
'x'},
 };
 
 /* printing format flags */
@@ -257,6 +261,7 @@ static const char 
commands_v_options[NUMBER_OF_CMD][NUMBER_OF_OPT] =
 #define FMT_PERSISTENTCONN     0x0020
 #define FMT_NOSORT             0x0040
 #define FMT_EXACT              0x0080
+#define FMT_GUE                0x0100
 
 #define SERVICE_NONE           0x0000
 #define SERVICE_ADDR           0x0001
@@ -300,6 +305,8 @@ enum {
        TAG_MCAST_PORT,
        TAG_MCAST_TTL,
        TAG_SYNC_MAXLEN,
+       TAG_GUE_PORT,
+       TAG_GUE,
 };
 
 /* various parsing helpers & parsing functions */
@@ -419,6 +426,7 @@ parse_options(int argc, char **argv, struct 
ipvs_command_entry *ce,
              unsigned int *options, unsigned int *format)
 {
        int c, parse;
+       u_int16_t gue_port;
        poptContext context;
        char *optarg = NULL, sched_flags_arg[128];
        struct poptOption options_table[] = {
@@ -495,6 +503,9 @@ parse_options(int argc, char **argv, struct 
ipvs_command_entry *ce,
                  TAG_MCAST_TTL, NULL, NULL },
                { "sync-maxlen", '\0', POPT_ARG_STRING, &optarg,
                  TAG_SYNC_MAXLEN, NULL, NULL },
+               { "gue-port", '\0', POPT_ARG_STRING, &optarg, TAG_GUE_PORT,
+                 NULL, NULL },
+               { "gue", '\0', POPT_ARG_NONE, NULL, TAG_GUE, NULL, NULL },
                { NULL, 0, 0, NULL, 0, NULL, NULL }
        };
 
@@ -773,6 +784,23 @@ parse_options(int argc, char **argv, struct 
ipvs_command_entry *ce,
                                fail(2, "illegal sync-maxlen specified");
                        ce->daemon.sync_maxlen = parse;
                        break;
+               case TAG_GUE_PORT:
+                       set_option(options, OPT_GUE_PORT);
+                       parse = string_to_number(optarg, 1, 65535);
+                       if (parse == -1)
+                               fail(2, "illegal gue port specified");
+                       gue_port = (u_int16_t) parse;
+                       ce->dest.conn_flags = (
+                                       (gue_port & 0xFFFE) << 16 |
+                                       (gue_port & 1) << 14 |
+                                       IP_VS_CONN_F_GUE |
+                                       ce->dest.conn_flags
+                       );
+                       break;
+               case TAG_GUE:
+                       set_option(options, OPT_GUE);
+                       *format |= FMT_GUE;
+                       break;
                default:
                        fail(2, "invalid option `%s'",
                             poptBadOption(context, POPT_BADOPTION_NOALIAS));
@@ -847,6 +875,7 @@ static int process_options(int argc, char **argv, int 
reading_stdin)
        struct ipvs_command_entry ce;
        unsigned int options = OPT_NONE;
        unsigned int format = FMT_NONE;
+       unsigned int fwd_method;
        int result = 0;
 
        memset(&ce, 0, sizeof(struct ipvs_command_entry));
@@ -883,6 +912,8 @@ static int process_options(int argc, char **argv, int 
reading_stdin)
        if (ce.cmd == CMD_STARTDAEMON && strlen(ce.daemon.mcast_ifn) == 0)
                strcpy(ce.daemon.mcast_ifn, DEF_MCAST_IFN);
 
+       fwd_method = ce.dest.conn_flags & IP_VS_CONN_F_FWD_MASK;
+
        if (ce.cmd == CMD_ADDDEST || ce.cmd == CMD_EDITDEST) {
                /*
                 * The destination port must be equal to the service port
@@ -890,15 +921,21 @@ static int process_options(int argc, char **argv, int 
reading_stdin)
                 * Don't worry about this if fwmark is used.
                 */
                if (!ce.svc.fwmark &&
-                   (ce.dest.conn_flags == IP_VS_CONN_F_TUNNEL
-                    || ce.dest.conn_flags == IP_VS_CONN_F_DROUTE))
+                   (fwd_method == IP_VS_CONN_F_TUNNEL
+                    || fwd_method == IP_VS_CONN_F_DROUTE))
                        ce.dest.port = ce.svc.port;
 
                /* Tunneling allows different address family */
                if (ce.dest.af != ce.svc.af &&
-                   ce.dest.conn_flags != IP_VS_CONN_F_TUNNEL)
+                   fwd_method != IP_VS_CONN_F_TUNNEL)
                        fail(2, "Different address family is allowed only "
                             "for tunneling servers");
+
+               /* Only tunneling allows gue encapsulation */
+               if ((options & OPT_GUE_PORT) &&
+                   fwd_method != IP_VS_CONN_F_TUNNEL)
+                       fail(2,
+                            "GUE encapsulation is allowed only for tunneling 
servers");
        }
 
        switch (ce.cmd) {
@@ -1301,6 +1338,7 @@ static void usage_exit(const char *program, const int 
exit_status)
                "  --gatewaying   -g                   gatewaying (direct 
routing) (default)\n"
                "  --ipip         -i                   ipip encapsulation 
(tunneling)\n"
                "  --masquerading -m                   masquerading (NAT)\n"
+               "  --gue-port      port                enable gue encapsulation 
with destination port\n"
                "  --weight       -w weight            capacity of real 
server\n"
                "  --u-threshold  -x uthreshold        upper threshold of 
connections\n"
                "  --l-threshold  -y lthreshold        lower threshold of 
connections\n"
@@ -1312,6 +1350,7 @@ static void usage_exit(const char *program, const int 
exit_status)
                "  --exact                             expand numbers (display 
exact values)\n"
                "  --thresholds                        output of thresholds 
information\n"
                "  --persistent-conn                   output of persistent 
connection info\n"
+               "  --gue                               output of gue 
encapsulation info (destination port)\n"
                "  --nosort                            disable sorting output 
of service/server entries\n"
                "  --sort                              does nothing, for 
backwards compatibility\n"
                "  --ops          -o                   one-packet scheduling\n"
@@ -1548,6 +1587,14 @@ static inline char *fwd_name(unsigned flags)
        return fwd;
 }
 
+static inline u_int16_t fwd_gue_port(unsigned int flags)
+{
+       u_int16_t gue_port;
+
+       gue_port = ((flags >> 16) & 0xFFFE) | ((flags >> 14) & 1);
+       return gue_port;
+}
+
 static inline char *fwd_switch(unsigned flags)
 {
        char *swt = NULL;
@@ -1641,6 +1688,9 @@ static void print_title(unsigned int format)
                       "  -> RemoteAddress:Port\n",
                       "Prot LocalAddress:Port",
                       "Weight", "PersistConn", "ActiveConn", "InActConn");
+       else if ((format & FMT_GUE))
+               printf("Prot LocalAddress:Port Scheduler Flags\n"
+                      "  -> RemoteAddress:Port           Forward GUE   Weight 
ActiveConn InActConn\n");
        else if (!(format & FMT_RULE))
                printf("Prot LocalAddress:Port Scheduler Flags\n"
                       "  -> RemoteAddress:Port           Forward Weight 
ActiveConn InActConn\n");
@@ -1778,8 +1828,20 @@ print_service_entry(ipvs_service_entry_t *se, unsigned 
int format)
                        dname[28] = '\0';
 
                if (format & FMT_RULE) {
-                       printf("-a %s -r %s %s -w %d\n", svc_name, dname,
-                              fwd_switch(e->conn_flags), e->weight);
+                       if (e->conn_flags & IP_VS_CONN_F_GUE) {
+                               printf("-a %s -r %s %s -w %d --gue-port %d\n",
+                                      svc_name,
+                                      dname,
+                                      fwd_switch(e->conn_flags),
+                                      e->weight,
+                                      fwd_gue_port(e->conn_flags));
+                       } else {
+                               printf("-a %s -r %s %s -w %d\n",
+                                      svc_name,
+                                      dname,
+                                      fwd_switch(e->conn_flags),
+                                      e->weight);
+                       }
                } else if (format & FMT_STATS) {
                        printf("  -> %-28s", dname);
                        print_largenum(e->stats64.conns, format);
@@ -1804,6 +1866,11 @@ print_service_entry(ipvs_service_entry_t *se, unsigned 
int format)
                        printf("  -> %-28s %-9u %-11u %-10u %-10u\n", dname,
                               e->weight, e->persistconns,
                               e->activeconns, e->inactconns);
+               } else if (format & FMT_GUE) {
+                       printf("  -> %-28s %-7s %-5d %-6d %-10u %-10u\n",
+                              dname, fwd_name(e->conn_flags),
+                              fwd_gue_port(e->conn_flags),
+                              e->weight, e->activeconns, e->inactconns);
                } else
                        printf("  -> %-28s %-7s %-6d %-10u %-10u\n",
                               dname, fwd_name(e->conn_flags),
diff --git a/libipvs/ip_vs.h b/libipvs/ip_vs.h
index ad0141c..6c7f6ee 100644
--- a/libipvs/ip_vs.h
+++ b/libipvs/ip_vs.h
@@ -101,6 +101,22 @@
 #define IP_VS_CONN_F_TEMPLATE  0x1000          /* template, not connection */
 #define IP_VS_CONN_F_ONE_PACKET        0x2000          /* forward only one 
packet */
 
+/* tunneling with gue */
+#define IP_VS_CONN_F_GUE       0x0008
+
+/* gue destination port bit 0 */
+#define IP_VS_CONN_F_GUE_PORT_LSB      0x4000
+
+/* gue destination port bit 1-15 */
+#define IP_VS_CONN_F_GUE_PORT_MSB      0xFFFFE0000
+
+/* Connection flags from destination that can be changed by user space */
+#define IP_VS_CONN_F_DEST_MASK (IP_VS_CONN_F_FWD_MASK | \
+                               IP_VS_CONN_F_GUE | \
+                               IP_VS_CONN_F_GUE_PORT_MSB | \
+                               IP_VS_CONN_F_GUE_PORT_LSB | \
+                               0)
+
 #define IP_VS_SCHEDNAME_MAXLEN 16
 #define IP_VS_PENAME_MAXLEN    16
 #define IP_VS_IFNAME_MAXLEN    16
diff --git a/libipvs/libipvs.c b/libipvs/libipvs.c
index 9be7700..a2d0018 100644
--- a/libipvs/libipvs.c
+++ b/libipvs/libipvs.c
@@ -388,7 +388,8 @@ static int ipvs_nl_fill_dest_attr(struct nl_msg *msg, 
ipvs_dest_t *dst)
        NLA_PUT_U16(msg, IPVS_DEST_ATTR_ADDR_FAMILY, dst->af);
        NLA_PUT(msg, IPVS_DEST_ATTR_ADDR, sizeof(dst->addr), &(dst->addr));
        NLA_PUT_U16(msg, IPVS_DEST_ATTR_PORT, dst->port);
-       NLA_PUT_U32(msg, IPVS_DEST_ATTR_FWD_METHOD, dst->conn_flags & 
IP_VS_CONN_F_FWD_MASK);
+       NLA_PUT_U32(msg, IPVS_DEST_ATTR_FWD_METHOD,
+                   dst->conn_flags & IP_VS_CONN_F_DEST_MASK);
        NLA_PUT_U32(msg, IPVS_DEST_ATTR_WEIGHT, dst->weight);
        NLA_PUT_U32(msg, IPVS_DEST_ATTR_U_THRESH, dst->u_threshold);
        NLA_PUT_U32(msg, IPVS_DEST_ATTR_L_THRESH, dst->l_threshold);
-- 
2.21.0