[PATCH v5 11/13] dhcp, dhcpv6: Select addresses for DHCP distribution

[Jon Maloy <jmaloy@redhat.com>]

We update DHCP to select and mark the best address to advertise among
the potentially multiple addresses in the address array.

We also extend DHCPv6 to advertise all suitable IPv6 addresses
(excluding observed and link-local addresses) in a single reply
message, per RFC 8415.

Finally, we update conf_print() to reflect above changes.

Signed-off-by: Jon Maloy <jmaloy@redhat.com>

---
v5: - Replaced fwd_first_usable_addr() with new helper functions
      to support these changes.
    - Broke out NDP part to a separate commit
---
 conf.c   | 73 +++++++++++++++++++++++++++++++++++-------------------
 conf.h   |  1 +
 dhcp.c   | 25 ++++++++++++-------
 dhcp.h   |  2 +-
 dhcpv6.c | 75 ++++++++++++++++++++++++++++++++++++++++++++------------
 5 files changed, 126 insertions(+), 50 deletions(-)

diff --git a/conf.c b/conf.c
index f622424..ea6dc07 100644
--- a/conf.c
+++ b/conf.c
@@ -47,6 +47,7 @@
 #include "lineread.h"
 #include "isolation.h"
 #include "log.h"
+#include "fwd.h"
 #include "vhost_user.h"
 
 #define NETNS_RUN_DIR	"/run/netns"
@@ -1116,11 +1117,12 @@ enum passt_modes conf_mode(int argc, char *argv[])
 static void conf_print(const struct ctx *c)
 {
 	char buf4[INET_ADDRSTRLEN], buf6[INET6_ADDRSTRLEN];
-	char bufmac[ETH_ADDRSTRLEN], ifn[IFNAMSIZ];
-	struct inany_addr_entry *e;
+	char bufmac[ETH_ADDRSTRLEN];
 	int i;
 
 	if (c->ifi4 > 0 || c->ifi6 > 0) {
+		char ifn[IFNAMSIZ];
+
 		info("Template interface: %s%s%s%s%s",
 		     c->ifi4 > 0 ? if_indextoname(c->ifi4, ifn) : "",
 		     c->ifi4 > 0 ? " (IPv4)" : "",
@@ -1161,21 +1163,23 @@ static void conf_print(const struct ctx *c)
 			     inet_ntop(AF_INET, &c->ip4.map_host_loopback,
 				       buf4, sizeof(buf4)));
 
-		e = first_v4(c);
-		if (e && !c->no_dhcp) {
-			uint32_t mask;
-
-			mask = IN4_MASK(inany_prefix4(e));
-
-			info("DHCP:");
-			info("    assign: %s",
-			     inet_ntop(AF_INET, inany_v4(&e->addr),
-				       buf4, sizeof(buf4)));
-			info("    mask: %s",
-			     inet_ntop(AF_INET, &mask,        buf4, sizeof(buf4)));
-			info("    router: %s",
-			     inet_ntop(AF_INET, &c->ip4.guest_gw,
-				       buf4, sizeof(buf4)));
+		if (!c->no_dhcp) {
+			const struct inany_addr_entry *e;
+
+			e = fwd_select_addr(c, AF_INET, CONF_ADDR_USER,
+					    CONF_ADDR_HOST, 0, CONF_ADDR_OBSERVED);
+			if (e) {
+				uint32_t mask = IN4_MASK(inany_prefix4(e));
+
+				info("DHCP:");
+				info("    assign: %s",
+				     inany_ntop(&e->addr, buf4, sizeof(buf4)));
+				info("    mask: %s",
+				     inet_ntop(AF_INET, &mask, buf4, sizeof(buf4)));
+				info("    router: %s",
+				     inet_ntop(AF_INET, &c->ip4.guest_gw,
+					       buf4, sizeof(buf4)));
+			}
 		}
 
 		for (i = 0; i < ARRAY_SIZE(c->ip4.dns); i++) {
@@ -1209,14 +1213,33 @@ static void conf_print(const struct ctx *c)
 		else
 			goto dns6;
 
-		e = first_v6(c);
-		info("    assign: %s", !e ? "" :
-		     inet_ntop(AF_INET6, &e->addr.a6, buf6, sizeof(buf6)));
-		info("    router: %s",
-		     inet_ntop(AF_INET6, &c->ip6.guest_gw, buf6, sizeof(buf6)));
-		info("    our link-local: %s",
-		     inet_ntop(AF_INET6, &c->ip6.our_tap_ll,
-			       buf6, sizeof(buf6)));
+		{
+			int skip = CONF_ADDR_OBSERVED | CONF_ADDR_LINKLOCAL;
+			const struct inany_addr_entry *e;
+			int first = 1;
+
+			for_each_addr(e, c, AF_INET6) {
+				if (e->flags & skip)
+					continue;
+				/* NDP requires /64 for SLAAC */
+				if (!c->no_ndp && e->prefix_len != 64)
+					continue;
+
+				info("    %s: %s/%d",
+				     first ? "assign" : "      ",
+				     inany_ntop(&e->addr, buf6, sizeof(buf6)),
+				     e->prefix_len);
+				first = 0;
+			}
+			if (!first) {
+				info("    router: %s",
+				     inet_ntop(AF_INET6, &c->ip6.guest_gw,
+					       buf6, sizeof(buf6)));
+				info("    our link-local: %s",
+				     inet_ntop(AF_INET6, &c->ip6.our_tap_ll,
+					       buf6, sizeof(buf6)));
+			}
+		}
 
 dns6:
 		for (i = 0; i < ARRAY_SIZE(c->ip6.dns); i++) {
diff --git a/conf.h b/conf.h
index 8b10ac6..b7d5fd2 100644
--- a/conf.h
+++ b/conf.h
@@ -13,6 +13,7 @@
 #define CONF_ADDR_HOST		BIT(1)		/* From host interface */
 #define CONF_ADDR_LINKLOCAL	BIT(2)		/* Link-local address */
 #define CONF_ADDR_OBSERVED	BIT(3)		/* Seen in guest traffic */
+#define CONF_ADDR_DHCP		BIT(4)		/* IPv4: assigned via DHCP */
 
 enum passt_modes conf_mode(int argc, char *argv[]);
 void conf(struct ctx *c, int argc, char **argv);
diff --git a/dhcp.c b/dhcp.c
index af473ee..b7eeac1 100644
--- a/dhcp.c
+++ b/dhcp.c
@@ -31,6 +31,8 @@
 #include "passt.h"
 #include "tap.h"
 #include "log.h"
+#include "fwd.h"
+#include "conf.h"
 #include "dhcp.h"
 
 /**
@@ -300,23 +302,22 @@ static void opt_set_dns_search(const struct ctx *c, size_t max_len)
  *
  * Return: 0 if it's not a DHCP message, 1 if handled, -1 on failure
  */
-int dhcp(const struct ctx *c, struct iov_tail *data)
+int dhcp(struct ctx *c, struct iov_tail *data)
 {
-	struct inany_addr_entry *e = first_v4(c);
+	struct in_addr addr, mask, dst;
 	char macstr[ETH_ADDRSTRLEN];
+	struct inany_addr_entry *e;
 	size_t mlen, dlen, opt_len;
-	struct in_addr mask, dst;
+	struct udphdr uh_storage;
 	struct ethhdr eh_storage;
 	struct iphdr iph_storage;
-	struct udphdr uh_storage;
+	const struct udphdr *uh;
 	const struct ethhdr *eh;
 	const struct iphdr *iph;
-	const struct udphdr *uh;
 	struct msg m_storage;
 	struct msg const *m;
-	struct in_addr addr;
 	struct msg reply;
-	unsigned int i;
+	int i;
 
 	eh = IOV_REMOVE_HEADER(data, eh_storage);
 	iph = IOV_PEEK_HEADER(data, iph_storage);
@@ -346,7 +347,13 @@ int dhcp(const struct ctx *c, struct iov_tail *data)
 	    m->op != BOOTREQUEST)
 		return -1;
 
-	ASSERT(e);
+	/* Select address to offer */
+	e = fwd_select_addr(c, AF_INET, CONF_ADDR_DHCP,
+			    CONF_ADDR_USER, CONF_ADDR_HOST, CONF_ADDR_OBSERVED);
+	if (!e)
+		return -1;
+
+	e->flags |= CONF_ADDR_DHCP;
 	addr = *inany_v4(&e->addr);
 
 	reply.op		= BOOTREPLY;
@@ -409,7 +416,7 @@ int dhcp(const struct ctx *c, struct iov_tail *data)
 
 	info("    from %s", eth_ntop(m->chaddr, macstr, sizeof(macstr)));
 
-	mask.s_addr = IN4_MASK(inany_prefix4(e));
+	mask.s_addr = e ? IN4_MASK(inany_prefix4(e)) : 0;
 	memcpy(opts[1].s,  &mask,                sizeof(mask));
 	memcpy(opts[3].s,  &c->ip4.guest_gw,     sizeof(c->ip4.guest_gw));
 	memcpy(opts[54].s, &c->ip4.our_tap_addr, sizeof(c->ip4.our_tap_addr));
diff --git a/dhcp.h b/dhcp.h
index cd50c99..7326c7d 100644
--- a/dhcp.h
+++ b/dhcp.h
@@ -6,7 +6,7 @@
 #ifndef DHCP_H
 #define DHCP_H
 
-int dhcp(const struct ctx *c, struct iov_tail *data);
+int dhcp(struct ctx *c, struct iov_tail *data);
 void dhcp_init(void);
 
 #endif /* DHCP_H */
diff --git a/dhcpv6.c b/dhcpv6.c
index ba200e5..3bc2e6d 100644
--- a/dhcpv6.c
+++ b/dhcpv6.c
@@ -31,6 +31,8 @@
 #include "passt.h"
 #include "tap.h"
 #include "log.h"
+#include "fwd.h"
+#include "conf.h"
 
 /**
  * struct opt_hdr - DHCPv6 option header
@@ -207,7 +209,7 @@ struct msg_hdr {
  * @hdr:		DHCP message header
  * @server_id:		Server Identifier option
  * @ia_na:		Non-temporary Address option
- * @ia_addr:		Address for IA_NA
+ * @ia_addr:		Addresses for IA_NA (variable, up to INANY_MAX_ADDRS)
  * @client_id:		Client Identifier, variable length
  * @dns_servers:	DNS Recursive Name Server, here just for storage size
  * @dns_search:		Domain Search List, here just for storage size
@@ -218,7 +220,7 @@ static struct resp_t {
 
 	struct opt_server_id server_id;
 	struct opt_ia_na ia_na;
-	struct opt_ia_addr ia_addr;
+	struct opt_ia_addr ia_addr[INANY_MAX_ADDRS];
 	struct opt_client_id client_id;
 	struct opt_dns_servers dns_servers;
 	struct opt_dns_search dns_search;
@@ -227,15 +229,11 @@ static struct resp_t {
 	{ 0 },
 	SERVER_ID,
 
-	{ { OPT_IA_NA,		OPT_SIZE_CONV(sizeof(struct opt_ia_na) +
-					      sizeof(struct opt_ia_addr) -
-					      sizeof(struct opt_hdr)) },
+	{ { OPT_IA_NA,		0 },  /* Length set dynamically */
 	  1, (uint32_t)~0U, (uint32_t)~0U
 	},
 
-	{ { OPT_IAAADR,		OPT_SIZE(ia_addr) },
-	  IN6ADDR_ANY_INIT, (uint32_t)~0U, (uint32_t)~0U
-	},
+	{ { { 0 }, IN6ADDR_ANY_INIT, 0, 0 } },  /* IA_ADDRs filled dynamically */
 
 	{ { OPT_CLIENTID,	0, },
 	  { 0 }
@@ -539,6 +537,44 @@ static size_t dhcpv6_client_fqdn_fill(const struct iov_tail *data,
 	return offset + sizeof(struct opt_hdr) + opt_len;
 }
 
+/**
+ * dhcpv6_ia_addr_fill() - Fill IA_ADDR options for all suitable addresses
+ * @c:		Execution context
+ *
+ * Fills resp.ia_addr[] with all non-linklocal, non-observed addresses and
+ * updates resp.ia_na.hdr.l with the correct length.
+ *
+ * Return: number of addresses filled
+ */
+static int dhcpv6_ia_addr_fill(const struct ctx *c)
+{
+	int skip = CONF_ADDR_OBSERVED | CONF_ADDR_LINKLOCAL;
+	const struct inany_addr_entry *e;
+	int count = 0;
+
+	for_each_addr(e, c, AF_INET6) {
+		if (e->flags & skip)
+			continue;
+		if (count >= INANY_MAX_ADDRS)
+			break;
+
+		resp.ia_addr[count].hdr.t = OPT_IAAADR;
+		resp.ia_addr[count].hdr.l = htons(sizeof(struct opt_ia_addr) -
+						  sizeof(struct opt_hdr));
+		resp.ia_addr[count].addr = e->addr.a6;
+		resp.ia_addr[count].pref_lifetime = (uint32_t)~0U;
+		resp.ia_addr[count].valid_lifetime = (uint32_t)~0U;
+		count++;
+	}
+
+	/* Update IA_NA length: header fields + all IA_ADDRs */
+	resp.ia_na.hdr.l = htons(sizeof(struct opt_ia_na) -
+				 sizeof(struct opt_hdr) +
+				 count * sizeof(struct opt_ia_addr));
+
+	return count;
+}
+
 /**
  * dhcpv6() - Check if this is a DHCPv6 message, reply as needed
  * @c:		Execution context
@@ -552,7 +588,6 @@ int dhcpv6(struct ctx *c, struct iov_tail *data,
 	   const struct in6_addr *daddr)
 {
 	const struct opt_server_id *server_id = NULL;
-	struct inany_addr_entry *e = first_v6(c);
 	const struct opt_hdr *client_id = NULL;
 	/* The _storage variables can't be local to the blocks they're used in,
 	 * because IOV_*_HEADER() may return pointers to them which are
@@ -567,11 +602,14 @@ int dhcpv6(struct ctx *c, struct iov_tail *data,
 	struct opt_hdr client_id_storage;
 	/* cppcheck-suppress [variableScope,unmatchedSuppression] */
 	struct opt_ia_na ia_storage;
+	struct in6_addr *ia_addr = NULL;
+	struct in6_addr first_addr;
 	const struct in6_addr *src;
 	struct msg_hdr mh_storage;
 	const struct msg_hdr *mh;
 	struct udphdr uh_storage;
 	const struct udphdr *uh;
+	int addr_count;
 	size_t mlen, n;
 
 	uh = IOV_REMOVE_HEADER(data, uh_storage);
@@ -615,6 +653,12 @@ int dhcpv6(struct ctx *c, struct iov_tail *data,
 	if (ia && ntohs(ia->hdr.l) < MIN(OPT_VSIZE(ia_na), OPT_VSIZE(ia_ta)))
 		return -1;
 
+	/* Fill IA_ADDRs for all suitable addresses */
+	addr_count = dhcpv6_ia_addr_fill(c);
+	if (addr_count > 0) {
+		first_addr = resp.ia_addr[0].addr;
+		ia_addr = &first_addr;
+	}
 	resp.hdr.type = TYPE_REPLY;
 	switch (mh->type) {
 	case TYPE_REQUEST:
@@ -627,7 +671,7 @@ int dhcpv6(struct ctx *c, struct iov_tail *data,
 		if (mh->type == TYPE_CONFIRM && server_id)
 			return -1;
 
-		if (e && dhcpv6_ia_notonlink(data, &e->addr.a6)) {
+		if (ia_addr && dhcpv6_ia_notonlink(data, ia_addr)) {
 
 			dhcpv6_send_ia_notonlink(c, data, &client_id_base,
 						 ntohs(client_id->l), mh->xid);
@@ -668,12 +712,14 @@ int dhcpv6(struct ctx *c, struct iov_tail *data,
 	if (ia)
 		resp.ia_na.iaid = ((struct opt_ia_na *)ia)->iaid;
 
+	/* Client_id goes right after the used IA_ADDRs */
+	n = offsetof(struct resp_t, ia_addr) +
+	    addr_count * sizeof(struct opt_ia_addr);
 	iov_to_buf(&client_id_base.iov[0], client_id_base.cnt,
-		   client_id_base.off, &resp.client_id,
+		   client_id_base.off, (char *)&resp + n,
 		   ntohs(client_id->l) + sizeof(struct opt_hdr));
 
-	n = offsetof(struct resp_t, client_id) +
-	    sizeof(struct opt_hdr) + ntohs(client_id->l);
+	n += sizeof(struct opt_hdr) + ntohs(client_id->l);
 	n = dhcpv6_dns_fill(c, (char *)&resp, n);
 	n = dhcpv6_client_fqdn_fill(data, c, (char *)&resp, n);
 
@@ -704,6 +750,5 @@ void dhcpv6_init(const struct ctx *c)
 	memcpy(resp_not_on_link.server_id.duid_lladdr,
 	       c->our_tap_mac, sizeof(c->our_tap_mac));
 
-	if (first_v6(c))
-		resp.ia_addr.addr = first_v6(c)->addr.a6;
+	/* Address is set dynamically in dhcpv6() */
 }
-- 
2.52.0