Merge pull request #70 from danielinux/l3_dev

Added support for L3, non-ethernet interfaces

Author: gasbytes

.github/workflows/linux.yml

@@ -34,6 +34,13 @@ jobs:
           timeout --preserve-status 5m sudo ./build/test-evloop
           sudo killall tcpdump || true
 
+      - name: Run standalone "event loop" TUN test
+        timeout-minutes: 5
+        run: |
+          set -euo pipefail
+          timeout --preserve-status 5m sudo ./build/test-evloop-tun
+          sudo killall tcpdump || true
+
       - name: Run standalone "IPsec esp" test
         timeout-minutes: 7
         run: |

CMakeLists.txt

@@ -93,6 +93,15 @@ add_executable(test-evloop ${EXCLUDE_TEST_BINARY}
     )
 add_test(NAME evloop COMMAND test-evloop)
 
+if (CMAKE_SYSTEM_NAME STREQUAL "Linux")
+    add_executable(test-evloop-tun ${EXCLUDE_TEST_BINARY}
+        src/test/test_eventloop_tun.c
+        src/port/posix/linux_tun.c
+        ${WOLFIP_SRCS}
+        )
+    add_test(NAME evloop-tun COMMAND test-evloop-tun)
+endif()
+
 add_executable(test-dns ${EXCLUDE_TEST_BINARY}
     src/test/test_dhcp_dns.c
     ${WOLFIP_SRCS}

Makefile

@@ -156,6 +156,9 @@ EXE=build/tcpecho build/tcp_netcat_poll build/tcp_netcat_select \
 	build/test-evloop build/test-dns build/test-wolfssl-forwarding \
 	build/test-ttl-expired build/test-wolfssl build/test-httpd \
 	build/ipfilter-logger build/test-esp build/esp-server
+ifeq ($(UNAME_S),Linux)
+  EXE+= build/test-evloop-tun
+endif
 LIB=libwolfip.so
 
 PREFIX=/usr/local
@@ -217,6 +220,10 @@ build/test-evloop: $(OBJ) build/test/test_eventloop.o
 	@echo "[LD] $@"
 	@$(CC) $(CFLAGS) -o $@ $(BEGIN_GROUP) $(^) $(LDFLAGS) $(END_GROUP)
 
+build/test-evloop-tun: $(OBJ) build/test/test_eventloop_tun.o build/port/posix/linux_tun.o
+	@echo "[LD] $@"
+	@$(CC) $(CFLAGS) -o $@ $(BEGIN_GROUP) $(^) $(LDFLAGS) $(END_GROUP)
+
 build/test-dns: $(OBJ) build/test/test_dhcp_dns.o
 	@echo "[LD] $@"
 	@$(CC) $(CFLAGS) -o $@ $(BEGIN_GROUP) $(^) $(LDFLAGS) $(END_GROUP)

docs/API.md

@@ -26,12 +26,15 @@ wolfIP is a minimal TCP/IP stack designed for resource-constrained embedded syst
 struct wolfIP_ll_dev {
     uint8_t mac[6];          // Device MAC address
     char ifname[16];         // Interface name
+    uint8_t non_ethernet;    // L3-only link (no Ethernet header/ARP when set)
     int (*poll)(struct wolfIP_ll_dev *ll, void *buf, uint32_t len);  // Receive function
     int (*send)(struct wolfIP_ll_dev *ll, void *buf, uint32_t len);  // Transmit function
 };
 ```
 wolfIP maintains an array of these descriptors sized by `WOLFIP_MAX_INTERFACES` (default `1`). Call `wolfIP_getdev_ex()` to access a specific slot; the legacy `wolfIP_getdev()` helper targets the first hardware slot (index `0` normally, or `1` when the optional loopback interface is enabled).
 
+When `non_ethernet` is set, the interface is treated as L3-only point-to-point: the stack skips ARP/neighbor resolution, omits Ethernet headers on transmit, and expects receive buffers to begin at the IP header.
+
 ### IP Configuration
 ```c
 struct ipconf {

src/port/posix/linux_tun.c

@@ -0,0 +1,237 @@
+/* linux_tun.c
+ *
+ * Linux TUN (L3) interface for wolfIP.
+ *
+ * Copyright (C) 2026 wolfSSL Inc.
+ *
+ * This file is part of wolfIP TCP/IP stack.
+ *
+ * wolfIP is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * wolfIP is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1335, USA
+ */
+#include <arpa/inet.h>
+#include <fcntl.h>
+#include <linux/if_tun.h>
+#include <net/if.h>
+#include <netinet/in.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/poll.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <unistd.h>
+#include <linux/netlink.h>
+#include <linux/rtnetlink.h>
+
+#define WOLF_POSIX
+#include "config.h"
+#include "wolfip.h"
+#undef WOLF_POSIX
+
+static int tun_fd = -1;
+
+static int tun_poll(struct wolfIP_ll_dev *ll, void *buf, uint32_t len)
+{
+    struct pollfd pfd;
+    int ret;
+    (void)ll;
+    if (tun_fd < 0)
+        return -1;
+    pfd.fd = tun_fd;
+    pfd.events = POLLIN;
+    ret = poll(&pfd, 1, 2);
+    if (ret < 0) {
+        perror("poll");
+        return -1;
+    }
+    if (ret == 0)
+        return 0;
+    return read(tun_fd, buf, len);
+}
+
+static int tun_send(struct wolfIP_ll_dev *ll, void *buf, uint32_t len)
+{
+    (void)ll;
+    if (tun_fd < 0)
+        return -1;
+    return write(tun_fd, buf, len);
+}
+
+static int nl_addattr(struct nlmsghdr *nlh, size_t maxlen, int type,
+                      const void *data, size_t alen)
+{
+    size_t len = RTA_LENGTH(alen);
+    struct rtattr *rta;
+    size_t newlen = NLMSG_ALIGN(nlh->nlmsg_len) + RTA_ALIGN(len);
+
+    if (newlen > maxlen)
+        return -1;
+    rta = (struct rtattr *)((char *)nlh + NLMSG_ALIGN(nlh->nlmsg_len));
+    rta->rta_type = type;
+    rta->rta_len = (unsigned short)len;
+    if (alen > 0 && data)
+        memcpy(RTA_DATA(rta), data, alen);
+    nlh->nlmsg_len = (unsigned int)newlen;
+    return 0;
+}
+
+static int tun_add_host_route(const char *ifname, uint32_t peer_ip)
+{
+    int fd;
+    int ifindex;
+    struct {
+        struct nlmsghdr nlh;
+        struct rtmsg rtm;
+        char buf[128];
+    } req;
+    struct sockaddr_nl nladdr;
+
+    ifindex = if_nametoindex(ifname);
+    if (ifindex == 0)
+        return -1;
+
+    memset(&req, 0, sizeof(req));
+    req.nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
+    req.nlh.nlmsg_type = RTM_NEWROUTE;
+    req.nlh.nlmsg_flags = NLM_F_REQUEST | NLM_F_CREATE | NLM_F_REPLACE;
+    req.nlh.nlmsg_seq = 1;
+    req.rtm.rtm_family = AF_INET;
+    req.rtm.rtm_dst_len = 32;
+    req.rtm.rtm_table = RT_TABLE_MAIN;
+    req.rtm.rtm_protocol = RTPROT_BOOT;
+    req.rtm.rtm_scope = RT_SCOPE_LINK;
+    req.rtm.rtm_type = RTN_UNICAST;
+
+    if (nl_addattr(&req.nlh, sizeof(req), RTA_DST, &peer_ip, sizeof(peer_ip)) < 0)
+        return -1;
+    if (nl_addattr(&req.nlh, sizeof(req), RTA_OIF, &ifindex, sizeof(ifindex)) < 0)
+        return -1;
+
+    fd = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+    if (fd < 0)
+        return -1;
+    memset(&nladdr, 0, sizeof(nladdr));
+    nladdr.nl_family = AF_NETLINK;
+    if (sendto(fd, &req, req.nlh.nlmsg_len, 0,
+               (struct sockaddr *)&nladdr, sizeof(nladdr)) < 0) {
+        close(fd);
+        return -1;
+    }
+    close(fd);
+    return 0;
+}
+
+int tun_init(struct wolfIP_ll_dev *ll, const char *ifname,
+             uint32_t host_ip, uint32_t peer_ip)
+{
+    struct ifreq ifr;
+    struct sockaddr_in *addr;
+    int sock_fd;
+
+    memset(&ifr, 0, sizeof(ifr));
+    ifr.ifr_flags = IFF_TUN | IFF_NO_PI;
+    strncpy(ifr.ifr_name, ifname, IFNAMSIZ);
+    ifr.ifr_name[IFNAMSIZ - 1] = '\0';
+
+    tun_fd = open("/dev/net/tun", O_RDWR);
+    if (tun_fd < 0 || ioctl(tun_fd, TUNSETIFF, (void *)&ifr) != 0) {
+        perror("ioctl TUNSETIFF");
+        if (tun_fd >= 0) {
+            close(tun_fd);
+            tun_fd = -1;
+        }
+        return -1;
+    }
+    {
+        int flags = fcntl(tun_fd, F_GETFL, 0);
+        if (flags >= 0)
+            (void)fcntl(tun_fd, F_SETFL, flags | O_NONBLOCK);
+    }
+
+    if (ll) {
+        memset(ll->mac, 0, sizeof(ll->mac));
+        strncpy(ll->ifname, ifname, sizeof(ll->ifname) - 1);
+        ll->ifname[sizeof(ll->ifname) - 1] = '\0';
+        ll->non_ethernet = 1;
+        ll->poll = tun_poll;
+        ll->send = tun_send;
+    }
+
+    sock_fd = socket(AF_INET, SOCK_DGRAM, 0);
+    if (sock_fd < 0) {
+        perror("socket");
+        close(tun_fd);
+        tun_fd = -1;
+        return -1;
+    }
+
+    memset(&ifr, 0, sizeof(ifr));
+    strncpy(ifr.ifr_name, ifname, IFNAMSIZ);
+    ifr.ifr_name[IFNAMSIZ - 1] = '\0';
+    if (ioctl(sock_fd, SIOCGIFFLAGS, &ifr) < 0) {
+        perror("ioctl SIOCGIFFLAGS");
+        close(sock_fd);
+        close(tun_fd);
+        tun_fd = -1;
+        return -1;
+    }
+    ifr.ifr_flags |= (IFF_UP | IFF_RUNNING | IFF_POINTOPOINT);
+    if (ioctl(sock_fd, SIOCSIFFLAGS, &ifr) < 0) {
+        perror("ioctl SIOCSIFFLAGS");
+        close(sock_fd);
+        close(tun_fd);
+        tun_fd = -1;
+        return -1;
+    }
+
+    addr = (struct sockaddr_in *)&ifr.ifr_addr;
+    addr->sin_family = AF_INET;
+    addr->sin_addr.s_addr = host_ip;
+    if (ioctl(sock_fd, SIOCSIFADDR, &ifr) < 0) {
+        perror("ioctl SIOCSIFADDR");
+        close(sock_fd);
+        close(tun_fd);
+        tun_fd = -1;
+        return -1;
+    }
+
+    addr = (struct sockaddr_in *)&ifr.ifr_dstaddr;
+    addr->sin_family = AF_INET;
+    addr->sin_addr.s_addr = peer_ip;
+    if (ioctl(sock_fd, SIOCSIFDSTADDR, &ifr) < 0) {
+        perror("ioctl SIOCSIFDSTADDR");
+        close(sock_fd);
+        close(tun_fd);
+        tun_fd = -1;
+        return -1;
+    }
+
+    addr = (struct sockaddr_in *)&ifr.ifr_netmask;
+    addr->sin_family = AF_INET;
+    addr->sin_addr.s_addr = htonl(0xFFFFFFFFU);
+    if (ioctl(sock_fd, SIOCSIFNETMASK, &ifr) < 0) {
+        perror("ioctl SIOCSIFNETMASK");
+        close(sock_fd);
+        close(tun_fd);
+        tun_fd = -1;
+        return -1;
+    }
+
+    (void)tun_add_host_route(ifname, peer_ip);
+    printf("Successfully initialized tun device %s\n", ifname);
+    close(sock_fd);
+    return 0;
+}