--- /dev/null
+------------------------------------------
+ Link-local IP address auto-configuration
+------------------------------------------
+
+Negotiate with other link-local clients on the local network
+for an address that doesn't require explicit configuration.
+This is especially useful if a DHCP server cannot be guaranteed
+to exist in all environments that the device must operate.
+
+This is an implementation of RFC3927.
+
+----------
+ Commands
+----------
+
+When CONFIG_CMD_LINK_LOCAL is defined in the board config file,
+the "linklocal" command is available. This running this will
+take approximately 5 seconds while the address is negotiated.
+
+------------------------
+ Environment interation
+------------------------
+
+The "llipaddr" variable is set with the most recently
+negotiated address and is preferred in future negotiations.
+
+The "ipaddr", "netmask", and "gatewayip" variables are set
+after successful negotiation to enable network access.
+
+-------------
+ Limitations
+-------------
+
+RFC3927 requires that addresses are continuously checked to
+avoid conflicts, however this can only happen when the NetLoop
+is getting called. It is possible for a conflict to go undetected
+until a command that accesses the network is executed.
+
+Using NetConsole is one way to ensure that NetLoop is always
+processing packets and monitoring for conflicts.
+
+This is also not a concern if the feature is use to connect
+directly to another machine that may not be running a DHCP server.
+
+----------------
+ Example script
+----------------
+
+This script allows use of DHCP and/or Link-local controlled
+by env variables. It depends on CONFIG_CMD_LINK_LOCAL, CONFIG_CMD_DHCP,
+and CONFIG_BOOTP_MAY_FAIL.
+If both fail or are disabled, static settings are used.
+
+#define CONFIG_EXTRA_ENV_SETTINGS \
+ "ipconfigcmd=if test \\\"$dhcpenabled\\\" -ne 0;" \
+ "then " \
+ "dhcpfail=0;dhcp || dhcpfail=1;" \
+ "else " \
+ "dhcpfail=-1;" \
+ "fi;" \
+ "if test \\\"$linklocalenabled\\\" -ne 0 -a " \
+ "\\\"$dhcpfail\\\" -ne 0;" \
+ "then " \
+ "linklocal;" \
+ "llfail=0;" \
+ "else " \
+ "llfail=-1;" \
+ "fi;" \
+ "if test \\\"$llfail\\\" -ne 0 -a " \
+ "\\\"$dhcpfail\\\" -ne 0; " \
+ "then " \
+ "setenv ipaddr $sipaddr; " \
+ "setenv netmask $snetmask; " \
+ "setenv gatewayip $sgatewayip; " \
+ "fi;\0" \
+
--- /dev/null
+/*
+ * RFC3927 ZeroConf IPv4 Link-Local addressing
+ * (see <http://www.zeroconf.org/>)
+ *
+ * Copied from BusyBox - networking/zcip.c
+ *
+ * Copyright (C) 2003 by Arthur van Hoff (avh@strangeberry.com)
+ * Copyright (C) 2004 by David Brownell
+ * Copyright (C) 2010 by Joe Hershberger
+ *
+ * Licensed under the GPL v2 or later
+ */
+
+#include <common.h>
+#include <net.h>
+#include "arp.h"
+#include "net_rand.h"
+
+/* We don't need more than 32 bits of the counter */
+#define MONOTONIC_MS() ((unsigned)get_timer(0) * (1000 / CONFIG_SYS_HZ))
+
+enum {
+/* 169.254.0.0 */
+ LINKLOCAL_ADDR = 0xa9fe0000,
+
+ IN_CLASSB_NET = 0xffff0000,
+ IN_CLASSB_HOST = 0x0000ffff,
+
+/* protocol timeout parameters, specified in seconds */
+ PROBE_WAIT = 1,
+ PROBE_MIN = 1,
+ PROBE_MAX = 2,
+ PROBE_NUM = 3,
+ MAX_CONFLICTS = 10,
+ RATE_LIMIT_INTERVAL = 60,
+ ANNOUNCE_WAIT = 2,
+ ANNOUNCE_NUM = 2,
+ ANNOUNCE_INTERVAL = 2,
+ DEFEND_INTERVAL = 10
+};
+
+/* States during the configuration process. */
+static enum ll_state_t {
+ PROBE = 0,
+ RATE_LIMIT_PROBE,
+ ANNOUNCE,
+ MONITOR,
+ DEFEND,
+ DISABLED
+} state = DISABLED;
+
+static IPaddr_t ip;
+static int timeout_ms = -1;
+static unsigned deadline_ms;
+static unsigned conflicts;
+static unsigned nprobes;
+static unsigned nclaims;
+static int ready;
+
+static void link_local_timeout(void);
+
+/**
+ * Pick a random link local IP address on 169.254/16, except that
+ * the first and last 256 addresses are reserved.
+ */
+static IPaddr_t pick(void)
+{
+ unsigned tmp;
+
+ do {
+ tmp = rand() & IN_CLASSB_HOST;
+ } while (tmp > (IN_CLASSB_HOST - 0x0200));
+ return (IPaddr_t) htonl((LINKLOCAL_ADDR + 0x0100) + tmp);
+}
+
+/**
+ * Return milliseconds of random delay, up to "secs" seconds.
+ */
+static inline unsigned random_delay_ms(unsigned secs)
+{
+ return rand() % (secs * 1000);
+}
+
+static void configure_wait(void)
+{
+ if (timeout_ms == -1)
+ return;
+
+ /* poll, being ready to adjust current timeout */
+ if (!timeout_ms)
+ timeout_ms = random_delay_ms(PROBE_WAIT);
+
+ /* set deadline_ms to the point in time when we timeout */
+ deadline_ms = MONOTONIC_MS() + timeout_ms;
+
+ debug("...wait %d %s nprobes=%u, nclaims=%u\n",
+ timeout_ms, eth_get_name(), nprobes, nclaims);
+
+ NetSetTimeout(timeout_ms, link_local_timeout);
+}
+
+void link_local_start(void)
+{
+ ip = getenv_IPaddr("llipaddr");
+ if (ip != 0 && (ip & IN_CLASSB_NET) != LINKLOCAL_ADDR) {
+ puts("invalid link address");
+ net_set_state(NETLOOP_FAIL);
+ return;
+ }
+ NetOurSubnetMask = IN_CLASSB_NET;
+
+ srand_mac();
+ if (ip == 0)
+ ip = pick();
+
+ state = PROBE;
+ timeout_ms = 0;
+ conflicts = 0;
+ nprobes = 0;
+ nclaims = 0;
+ ready = 0;
+
+ configure_wait();
+}
+
+static void link_local_timeout(void)
+{
+ switch (state) {
+ case PROBE:
+ /* timeouts in the PROBE state mean no conflicting ARP packets
+ have been received, so we can progress through the states */
+ if (nprobes < PROBE_NUM) {
+ nprobes++;
+ debug("probe/%u %s@%pI4\n",
+ nprobes, eth_get_name(), &ip);
+ arp_raw_request(0, NetEtherNullAddr, ip);
+ timeout_ms = PROBE_MIN * 1000;
+ timeout_ms += random_delay_ms(PROBE_MAX - PROBE_MIN);
+ } else {
+ /* Switch to announce state */
+ state = ANNOUNCE;
+ nclaims = 0;
+ debug("announce/%u %s@%pI4\n",
+ nclaims, eth_get_name(), &ip);
+ arp_raw_request(ip, NetOurEther, ip);
+ timeout_ms = ANNOUNCE_INTERVAL * 1000;
+ }
+ break;
+ case RATE_LIMIT_PROBE:
+ /* timeouts in the RATE_LIMIT_PROBE state mean no conflicting
+ ARP packets have been received, so we can move immediately
+ to the announce state */
+ state = ANNOUNCE;
+ nclaims = 0;
+ debug("announce/%u %s@%pI4\n",
+ nclaims, eth_get_name(), &ip);
+ arp_raw_request(ip, NetOurEther, ip);
+ timeout_ms = ANNOUNCE_INTERVAL * 1000;
+ break;
+ case ANNOUNCE:
+ /* timeouts in the ANNOUNCE state mean no conflicting ARP
+ packets have been received, so we can progress through
+ the states */
+ if (nclaims < ANNOUNCE_NUM) {
+ nclaims++;
+ debug("announce/%u %s@%pI4\n",
+ nclaims, eth_get_name(), &ip);
+ arp_raw_request(ip, NetOurEther, ip);
+ timeout_ms = ANNOUNCE_INTERVAL * 1000;
+ } else {
+ /* Switch to monitor state */
+ state = MONITOR;
+ printf("Successfully assigned %pI4\n", &ip);
+ NetCopyIP(&NetOurIP, &ip);
+ ready = 1;
+ conflicts = 0;
+ timeout_ms = -1;
+ /* Never timeout in the monitor state */
+ NetSetTimeout(0, NULL);
+
+ /* NOTE: all other exit paths should deconfig ... */
+ net_set_state(NETLOOP_SUCCESS);
+ return;
+ }
+ break;
+ case DEFEND:
+ /* We won! No ARP replies, so just go back to monitor */
+ state = MONITOR;
+ timeout_ms = -1;
+ conflicts = 0;
+ break;
+ default:
+ /* Invalid, should never happen. Restart the whole protocol */
+ state = PROBE;
+ ip = pick();
+ timeout_ms = 0;
+ nprobes = 0;
+ nclaims = 0;
+ break;
+ }
+ configure_wait();
+}
+
+void link_local_receive_arp(struct arp_hdr *arp, int len)
+{
+ int source_ip_conflict;
+ int target_ip_conflict;
+
+ if (state == DISABLED)
+ return;
+
+ /* We need to adjust the timeout in case we didn't receive a
+ conflicting packet. */
+ if (timeout_ms > 0) {
+ unsigned diff = deadline_ms - MONOTONIC_MS();
+ if ((int)(diff) < 0) {
+ /* Current time is greater than the expected timeout
+ time. This should never happen */
+ debug("missed an expected timeout\n");
+ timeout_ms = 0;
+ } else {
+ debug("adjusting timeout\n");
+ timeout_ms = diff | 1; /* never 0 */
+ }
+ }
+/*
+ * XXX Don't bother with ethernet link just yet
+ if ((fds[0].revents & POLLIN) == 0) {
+ if (fds[0].revents & POLLERR) {
+ // FIXME: links routinely go down;
+ // this shouldn't necessarily exit.
+ bb_error_msg("iface %s is down", eth_get_name());
+ if (ready) {
+ run(argv, "deconfig", &ip);
+ }
+ return EXIT_FAILURE;
+ }
+ continue;
+ }
+*/
+
+ debug("%s recv arp type=%d, op=%d,\n",
+ eth_get_name(), ntohs(arp->ar_pro),
+ ntohs(arp->ar_op));
+ debug("\tsource=%pM %pI4\n",
+ &arp->ar_sha,
+ &arp->ar_spa);
+ debug("\ttarget=%pM %pI4\n",
+ &arp->ar_tha,
+ &arp->ar_tpa);
+
+ if (arp->ar_op != htons(ARPOP_REQUEST)
+ && arp->ar_op != htons(ARPOP_REPLY)
+ ) {
+ configure_wait();
+ return;
+ }
+
+ source_ip_conflict = 0;
+ target_ip_conflict = 0;
+
+ if (memcmp(&arp->ar_spa, &ip, ARP_PLEN) == 0
+ && memcmp(&arp->ar_sha, NetOurEther, ARP_HLEN) != 0
+ ) {
+ source_ip_conflict = 1;
+ }
+ if (arp->ar_op == htons(ARPOP_REQUEST)
+ && memcmp(&arp->ar_tpa, &ip, ARP_PLEN) == 0
+ && memcmp(&arp->ar_tha, NetOurEther, ARP_HLEN) != 0
+ ) {
+ target_ip_conflict = 1;
+ }
+
+ debug("state = %d, source ip conflict = %d, target ip conflict = %d\n",
+ state, source_ip_conflict, target_ip_conflict);
+ switch (state) {
+ case PROBE:
+ case ANNOUNCE:
+ /* When probing or announcing, check for source IP conflicts
+ and other hosts doing ARP probes (target IP conflicts). */
+ if (source_ip_conflict || target_ip_conflict) {
+ conflicts++;
+ state = PROBE;
+ if (conflicts >= MAX_CONFLICTS) {
+ debug("%s ratelimit\n", eth_get_name());
+ timeout_ms = RATE_LIMIT_INTERVAL * 1000;
+ state = RATE_LIMIT_PROBE;
+ }
+
+ /* restart the whole protocol */
+ ip = pick();
+ timeout_ms = 0;
+ nprobes = 0;
+ nclaims = 0;
+ }
+ break;
+ case MONITOR:
+ /* If a conflict, we try to defend with a single ARP probe */
+ if (source_ip_conflict) {
+ debug("monitor conflict -- defending\n");
+ state = DEFEND;
+ timeout_ms = DEFEND_INTERVAL * 1000;
+ arp_raw_request(ip, NetOurEther, ip);
+ }
+ break;
+ case DEFEND:
+ /* Well, we tried. Start over (on conflict) */
+ if (source_ip_conflict) {
+ state = PROBE;
+ debug("defend conflict -- starting over\n");
+ ready = 0;
+ NetOurIP = 0;
+
+ /* restart the whole protocol */
+ ip = pick();
+ timeout_ms = 0;
+ nprobes = 0;
+ nclaims = 0;
+ }
+ break;
+ default:
+ /* Invalid, should never happen. Restart the whole protocol */
+ debug("invalid state -- starting over\n");
+ state = PROBE;
+ ip = pick();
+ timeout_ms = 0;
+ nprobes = 0;
+ nclaims = 0;
+ break;
+ }
+ configure_wait();
+}