Kconfig := src/target/Kconfig
.PHONY += all
-all: src/target/tuntest/firmly0cpm.bin
+all: bin/firmerware.bin
-%.o: %.c
- gcc -c -Iinclude -ggdb3 -o $@ $^
+#
+# Load configuration file and various Makefiles
+#
+-include .config
include bin/kconfig/Makefile
-include src/target/tuntest/Makefile
+include src/target/Makefile
+include src/driver/Makefile
include src/net/Makefile
+#
+# The main building target
+#
+bin/firmerware.bin: $(objs-y)
+ gcc -o $@ $^
+
.PHONY += clean
clean:
find . -name \*\.o -exec rm {} \;
rm -f src/target/tuntest/firmly0cpm.bin
+
+
+%.o: %.c
+ gcc -c -Iinclude -ggdb3 -o $@ $^
+
+.PHONY += help
+help:
+ @echo
+ @echo "*** Try 'make menuconfig' followed by 'make'"
+ @echo
+ @echo Platform-specific notes are part of the target help descriptions.
+ @echo
+++ /dev/null
-/* tuntest.c -- Tunnel-based testing of the network code.
- *
- * From: Rick van Rein <rick@openfortress.nl>
- */
-
-
-#include <stdlib.h>
-#include <stdio.h>
-#include <stdint.h>
-#include <stdarg.h>
-#include <string.h>
-#include <fcntl.h>
-
-#include <sys/types.h>
-#include <sys/ioctl.h>
-#include <sys/socket.h>
-#include <sys/select.h>
-
-#include <netinet/ip.h>
-#include <netinet/ip6.h>
-#include <netinet/udp.h>
-#include <netinet/ip_icmp.h>
-#include <netinet/icmp6.h>
-
-#include <linux/if.h>
-#include <linux/if_tun.h>
-#include <linux/if_ether.h>
-
-#include <0cpm/netfun.h>
-#include <0cpm/netdb.h>
-
-
-
-int tunsox = -1;
-
-struct tunbuf rbuf, wbuf;
-
-
-
-/*
- * TEST OUTPUT FUNCTIONS
- */
-
-uint8_t *net_arp_reply (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- printf ("Received: ARP Reply\n");
- return NULL;
-}
-
-uint8_t *net_rtp (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- printf ("Received: RTP\n");
- return NULL;
-}
-
-uint8_t *net_rtcp (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- printf ("Received: RTCP\n");
- return NULL;
-}
-
-uint8_t *net_sip (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- printf ("Received: SIP\n");
- return NULL;
-}
-
-uint8_t *net_mdns_resp_error (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- //TODO// printf ("Received: mDNS error response\n");
- return NULL;
-}
-
-uint8_t *net_mdns_resp_dyn (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- //TODO// printf ("Received: mDNS dynamic response\n");
- return NULL;
-}
-
-uint8_t *net_mdns_resp_std (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- //TODO// printf ("Received: mDNS standard response\n");
- return NULL;
-}
-
-uint8_t *net_mdns_query_error (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- //TODO// printf ("Received: mDNS error query\n");
- return NULL;
-}
-
-uint8_t *net_mdns_query_ok (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
- //TODO// printf ("Received: mDNS ok query\n");
- return NULL;
-}
-
-
-
-/* A collection of initialisation steps.
- */
-void init (void) {
- int i;
- for (i=0; i<IP6BINDING_COUNT; i++) {
- ip6binding [i].flags = I6B_EXPIRED;
- }
-}
-
-/*
- * TEST SETUP FUNCTIONS
- */
-
-typedef void *retfn (uint8_t *pout, uint32_t *mem);
-
-int process_packets (int sox, int secs) {
- int ok = 1;
- time_t maxtm = secs + time (NULL);
- while (ok) {
- fd_set seln;
- FD_ZERO (&seln);
- FD_SET (sox, &seln);
- struct timeval tout;
- tout.tv_sec = maxtm - time (NULL);
- tout.tv_usec = 500000;
- if ((tout.tv_sec < 0) || (select (sox+1, &seln, NULL, NULL, &tout) <= 0)) {
- return 1;
- }
- size_t len = read (sox, &rbuf, sizeof (struct tun_pi) + sizeof (rbuf.data));
- uint8_t *pkt = rbuf.data;
- uint32_t pktlen = len - sizeof (struct tun_pi);
- if (pktlen > 18) {
- uint32_t mem [28];
- bzero (mem, sizeof (mem));
- retfn *rf = (retfn *) netinput (rbuf.data, len - sizeof (struct tun_pi), mem);
-#if 0
- int i;
- for (i=0; i<28; i++) {
- printf (" M[%d]=0x%08x%s", i, mem [i],
- ((i+1)%4 == 0)? "\n": "");
- }
-#endif
- if (rf != NULL) {
- uint8_t *start = wbuf.data;
- uint8_t *stop = (*rf) (start, mem);
- if (stop) {
- mem [MEM_ALL_DONE] = (uint32_t) stop;
- uint32_t len = mem [MEM_ALL_DONE] - mem [MEM_ETHER_HEAD];
- if (netcore_send_buffer (tunsox, mem, &wbuf) < 0) {
- printf ("Failed while sending reply data\n");
- }
- }
- }
- }
- }
- return 0;
-}
-
-
-int setup_tunnel (void) {
- tunsox = -1;
- int ok = 1;
- //
- // Open the tunnel
- tunsox = open ("/dev/net/tun", O_RDWR);
- if (tunsox < 0) {
- ok = 0;
- }
- //
- // Set the tunnel name to "test0cpm"
- struct ifreq ifreq;
- bzero (&ifreq, sizeof (ifreq));
- strncpy (ifreq.ifr_name, "test0cpm", IFNAMSIZ);
- ifreq.ifr_flags = IFF_TAP;
- if (ok)
- if (ioctl (tunsox, TUNSETIFF, &ifreq) == -1) {
- ok = 0;
- }
- //
- // Set to promiscuous mode (meaningless for a tunnel)
-#if 0
- if (ok)
- if (ioctl(tunsox, SIOCGIFFLAGS, &ifreq) == -1) {
- ok = 0;
- }
- ifreq.ifr_flags |= IFF_PROMISC | IFF_UP;
- if (ok)
- if (ioctl(tunsox, SIOCSIFFLAGS, &ifreq) == -1) {
- ok = 0;
- }
-#endif
- //
- // Cleanup and report success or failure
- if (!ok) {
- if (tunsox >= 0) {
- close (tunsox);
- tunsox = -1;
- }
- return -1;
- }
- return tunsox;
-}
-
-
-
-int main (int argc, char *argv []) {
- init ();
- int tunsox = setup_tunnel ();
- if (tunsox == -1) {
- fprintf (stderr, "Failed to setup tunnel\n");
- exit (1);
- }
- system ("/sbin/ifconfig test0cpm up");
- system ("/sbin/ifconfig test0cpm");
- sleep (10);
- if (!netcore_bootstrap ()) {
- fprintf (stderr, "Failed to bootstrap the network\n");
- exit (1);
- }
- process_packets (tunsox, 60);
- printf ("Done.\n");
- close (tunsox);
- return 0;
-}
-
--- /dev/null
+/* tuntest.c -- Tunnel-based testing of the network code.
+ *
+ * From: Rick van Rein <rick@openfortress.nl>
+ */
+
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <stdarg.h>
+#include <string.h>
+#include <fcntl.h>
+
+#include <sys/types.h>
+#include <sys/ioctl.h>
+#include <sys/socket.h>
+#include <sys/select.h>
+
+#include <netinet/ip.h>
+#include <netinet/ip6.h>
+#include <netinet/udp.h>
+#include <netinet/ip_icmp.h>
+#include <netinet/icmp6.h>
+
+#include <linux/if.h>
+#include <linux/if_tun.h>
+#include <linux/if_ether.h>
+
+#include <0cpm/netfun.h>
+#include <0cpm/netdb.h>
+
+
+
+int tunsox = -1;
+
+struct tunbuf rbuf, wbuf;
+
+
+
+/*
+ * TEST OUTPUT FUNCTIONS
+ */
+
+uint8_t *net_arp_reply (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ printf ("Received: ARP Reply\n");
+ return NULL;
+}
+
+uint8_t *net_rtp (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ printf ("Received: RTP\n");
+ return NULL;
+}
+
+uint8_t *net_rtcp (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ printf ("Received: RTCP\n");
+ return NULL;
+}
+
+uint8_t *net_sip (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ printf ("Received: SIP\n");
+ return NULL;
+}
+
+uint8_t *net_mdns_resp_error (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ //TODO// printf ("Received: mDNS error response\n");
+ return NULL;
+}
+
+uint8_t *net_mdns_resp_dyn (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ //TODO// printf ("Received: mDNS dynamic response\n");
+ return NULL;
+}
+
+uint8_t *net_mdns_resp_std (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ //TODO// printf ("Received: mDNS standard response\n");
+ return NULL;
+}
+
+uint8_t *net_mdns_query_error (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ //TODO// printf ("Received: mDNS error query\n");
+ return NULL;
+}
+
+uint8_t *net_mdns_query_ok (uint8_t *pkt, uint32_t pktlen, uint32_t *mem) {
+ //TODO// printf ("Received: mDNS ok query\n");
+ return NULL;
+}
+
+
+
+/* A collection of initialisation steps.
+ */
+void init (void) {
+ int i;
+ for (i=0; i<IP6BINDING_COUNT; i++) {
+ ip6binding [i].flags = I6B_EXPIRED;
+ }
+}
+
+/*
+ * TEST SETUP FUNCTIONS
+ */
+
+typedef void *retfn (uint8_t *pout, uint32_t *mem);
+
+int process_packets (int sox, int secs) {
+ int ok = 1;
+ time_t maxtm = secs + time (NULL);
+ while (ok) {
+ fd_set seln;
+ FD_ZERO (&seln);
+ FD_SET (sox, &seln);
+ struct timeval tout;
+ tout.tv_sec = maxtm - time (NULL);
+ tout.tv_usec = 500000;
+ if ((tout.tv_sec < 0) || (select (sox+1, &seln, NULL, NULL, &tout) <= 0)) {
+ return 1;
+ }
+ size_t len = read (sox, &rbuf, sizeof (struct tun_pi) + sizeof (rbuf.data));
+ uint8_t *pkt = rbuf.data;
+ uint32_t pktlen = len - sizeof (struct tun_pi);
+ if (pktlen > 18) {
+ uint32_t mem [28];
+ bzero (mem, sizeof (mem));
+ retfn *rf = (retfn *) netinput (rbuf.data, len - sizeof (struct tun_pi), mem);
+#if 0
+ int i;
+ for (i=0; i<28; i++) {
+ printf (" M[%d]=0x%08x%s", i, mem [i],
+ ((i+1)%4 == 0)? "\n": "");
+ }
+#endif
+ if (rf != NULL) {
+ uint8_t *start = wbuf.data;
+ uint8_t *stop = (*rf) (start, mem);
+ if (stop) {
+ mem [MEM_ALL_DONE] = (uint32_t) stop;
+ uint32_t len = mem [MEM_ALL_DONE] - mem [MEM_ETHER_HEAD];
+ if (netcore_send_buffer (tunsox, mem, &wbuf) < 0) {
+ printf ("Failed while sending reply data\n");
+ }
+ }
+ }
+ }
+ }
+ return 0;
+}
+
+
+int setup_tunnel (void) {
+ tunsox = -1;
+ int ok = 1;
+ //
+ // Open the tunnel
+ tunsox = open ("/dev/net/tun", O_RDWR);
+ if (tunsox < 0) {
+ ok = 0;
+ }
+ //
+ // Set the tunnel name to "test0cpm"
+ struct ifreq ifreq;
+ bzero (&ifreq, sizeof (ifreq));
+ strncpy (ifreq.ifr_name, "test0cpm", IFNAMSIZ);
+ ifreq.ifr_flags = IFF_TAP;
+ if (ok)
+ if (ioctl (tunsox, TUNSETIFF, &ifreq) == -1) {
+ ok = 0;
+ }
+ //
+ // Set to promiscuous mode (meaningless for a tunnel)
+#if 0
+ if (ok)
+ if (ioctl(tunsox, SIOCGIFFLAGS, &ifreq) == -1) {
+ ok = 0;
+ }
+ ifreq.ifr_flags |= IFF_PROMISC | IFF_UP;
+ if (ok)
+ if (ioctl(tunsox, SIOCSIFFLAGS, &ifreq) == -1) {
+ ok = 0;
+ }
+#endif
+ //
+ // Cleanup and report success or failure
+ if (!ok) {
+ if (tunsox >= 0) {
+ close (tunsox);
+ tunsox = -1;
+ }
+ return -1;
+ }
+ return tunsox;
+}
+
+
+
+int main (int argc, char *argv []) {
+ init ();
+ int tunsox = setup_tunnel ();
+ if (tunsox == -1) {
+ fprintf (stderr, "Failed to setup tunnel\n");
+ exit (1);
+ }
+ system ("/sbin/ifconfig test0cpm up");
+ system ("/sbin/ifconfig test0cpm");
+ sleep (10);
+ if (!netcore_bootstrap ()) {
+ fprintf (stderr, "Failed to bootstrap the network\n");
+ exit (1);
+ }
+ process_packets (tunsox, 60);
+ printf ("Done.\n");
+ close (tunsox);
+ return 0;
+}
+
menu "Network Configuration"
-config NET_6BED4_V4ADDR
-
+config NET_6BED4_SERVER
+ string "6bed4 server IPv4 address"
+ help
+ Specify the IPv4 address of the remote tunnel endpoint
+ that will unpack 6bed4 packages. This is basically
+ IPv6 stateless autoconfiguration over an UDP-tunnel.
+
+ Details about 6bed4 can be found on:
+ http://devel.0cpm.org/6bed4/
+
+ Work is underway for making this globally available on
+ an anycast-address. Until that has been setup, there is
+ a need to specify an IPv4 address explicitly.
+
+endmenu
+
-src/net/net.o: src/net/core.o src/net/input.o src/net/reply.o src/net/send.o src/net/db.o
- ld -r -o $@ $^
+src/net/6bed4.c:
+ printf > $@ '#include <stdint.h>\nuint32_t ip4_6bed4 = ( %d << 24 | %d << 16 | %d << 8 | %d );\n' `echo $(CONFIG_NET_6BED4_SERVER) | sed 's/\./ /g'`
+
+objs-y += src/net/core.o src/net/input.o src/net/reply.o src/net/send.o src/net/db.o src/net/6bed4.o
uint8_t ether_zero [ETHER_ADDR_LEN] = { 0, 0, 0, 0, 0, 0 };
uint8_t prefix_6bed4 [8] = {
0x20, 0x01, 0xab, 0xba, 0x6b, 0xed, 0x00, 0x04 };
-uint32_t ip4_6bed4 = ( 123 << 24 | 45 << 16 | 67 << 8 | 89 );
+extern uint32_t ip4_6bed4;
uint8_t ip4_mine [4] = { 192, 168, 3, 13 };
extern uint8_t linklocal_mine [];
mainmenu "Configuration for firmly.0cpm SIP telephony"
-
choice
prompt "Target platform"
default TARGET_LINUX_RAWSOCK
TODO: Demo entry. Not implemented yet.
endchoice
+
+
+source "src/net/Kconfig"
--- /dev/null
+#
+# Sort each target's object files into objs-y or objs-n:
+#
+
+objs-$(CONFIG_TARGET_LINUX_TUNTEST) += src/driver/tuntest.o
+
+++ /dev/null
-src/target/tuntest/firmly0cpm.bin: src/driver/tunnel/tuntest.o src/net/net.o
- gcc -ggdb3 -o $@ $^
projects / firmerware / commitdiff