diff options
Diffstat (limited to 'tap.c')
| -rw-r--r-- | tap.c | 306 |
1 files changed, 201 insertions, 105 deletions
@@ -58,6 +58,8 @@ #include "packet.h" #include "tap.h" #include "log.h" +#include "vhost_user.h" +#include "vu_common.h" /* IPv4 (plus ARP) and IPv6 message batches from tap/guest to IP handlers */ static PACKET_POOL_NOINIT(pool_tap4, TAP_MSGS, pkt_buf); @@ -78,16 +80,22 @@ void tap_send_single(const struct ctx *c, const void *data, size_t l2len) struct iovec iov[2]; size_t iovcnt = 0; - if (c->mode == MODE_PASST) { + switch (c->mode) { + case MODE_PASST: iov[iovcnt] = IOV_OF_LVALUE(vnet_len); iovcnt++; - } - - iov[iovcnt].iov_base = (void *)data; - iov[iovcnt].iov_len = l2len; - iovcnt++; + /* fall through */ + case MODE_PASTA: + iov[iovcnt].iov_base = (void *)data; + iov[iovcnt].iov_len = l2len; + iovcnt++; - tap_send_frames(c, iov, iovcnt, 1); + tap_send_frames(c, iov, iovcnt, 1); + break; + case MODE_VU: + vu_send_single(c, data, l2len); + break; + } } /** @@ -118,8 +126,8 @@ static void *tap_push_l2h(const struct ctx *c, void *buf, uint16_t proto) struct ethhdr *eh = (struct ethhdr *)buf; /* TODO: ARP table lookup */ - memcpy(eh->h_dest, c->mac_guest, ETH_ALEN); - memcpy(eh->h_source, c->mac, ETH_ALEN); + memcpy(eh->h_dest, c->guest_mac, ETH_ALEN); + memcpy(eh->h_source, c->our_tap_mac, ETH_ALEN); eh->h_proto = ntohs(proto); return eh + 1; } @@ -172,11 +180,16 @@ void tap_udp4_send(const struct ctx *c, struct in_addr src, in_port_t sport, struct iphdr *ip4h = tap_push_l2h(c, buf, ETH_P_IP); struct udphdr *uh = tap_push_ip4h(ip4h, src, dst, l4len, IPPROTO_UDP); char *data = (char *)(uh + 1); + const struct iovec iov = { + .iov_base = (void *)in, + .iov_len = dlen + }; + struct iov_tail payload = IOV_TAIL(&iov, 1, 0); uh->source = htons(sport); uh->dest = htons(dport); uh->len = htons(l4len); - csum_udp4(uh, src, dst, in, dlen); + csum_udp4(uh, src, dst, &payload); memcpy(data, in, dlen); tap_send_single(c, buf, dlen + (data - buf)); @@ -247,7 +260,7 @@ static void *tap_push_ip6h(struct ipv6hdr *ip6h, void tap_udp6_send(const struct ctx *c, const struct in6_addr *src, in_port_t sport, const struct in6_addr *dst, in_port_t dport, - uint32_t flow, const void *in, size_t dlen) + uint32_t flow, void *in, size_t dlen) { size_t l4len = dlen + sizeof(struct udphdr); char buf[USHRT_MAX]; @@ -255,11 +268,16 @@ void tap_udp6_send(const struct ctx *c, struct udphdr *uh = tap_push_ip6h(ip6h, src, dst, l4len, IPPROTO_UDP, flow); char *data = (char *)(uh + 1); + const struct iovec iov = { + .iov_base = in, + .iov_len = dlen + }; + struct iov_tail payload = IOV_TAIL(&iov, 1, 0); uh->source = htons(sport); uh->dest = htons(dport); uh->len = htons(l4len); - csum_udp6(uh, src, dst, in, dlen); + csum_udp6(uh, src, dst, &payload); memcpy(data, in, dlen); tap_send_single(c, buf, dlen + (data - buf)); @@ -406,10 +424,18 @@ size_t tap_send_frames(const struct ctx *c, const struct iovec *iov, if (!nframes) return 0; - if (c->mode == MODE_PASTA) + switch (c->mode) { + case MODE_PASTA: m = tap_send_frames_pasta(c, iov, bufs_per_frame, nframes); - else + break; + case MODE_PASST: m = tap_send_frames_passt(c, iov, bufs_per_frame, nframes); + break; + case MODE_VU: + /* fall through */ + default: + ASSERT(0); + } if (m < nframes) debug("tap: failed to send %zu frames of %zu", @@ -795,6 +821,9 @@ resume: if (IN6_IS_ADDR_UNSPECIFIED(&c->ip6.addr_seen)) { c->ip6.addr_seen = *saddr; } + + if (IN6_IS_ADDR_UNSPECIFIED(&c->ip6.addr)) + c->ip6.addr = *saddr; } else if (!IN6_IS_ADDR_UNSPECIFIED(saddr)){ c->ip6.addr_seen = *saddr; } @@ -946,9 +975,9 @@ void tap_add_packet(struct ctx *c, ssize_t l2len, char *p) eh = (struct ethhdr *)p; - if (memcmp(c->mac_guest, eh->h_source, ETH_ALEN)) { - memcpy(c->mac_guest, eh->h_source, ETH_ALEN); - proto_update_l2_buf(c->mac_guest, NULL); + if (memcmp(c->guest_mac, eh->h_source, ETH_ALEN)) { + memcpy(c->guest_mac, eh->h_source, ETH_ALEN); + proto_update_l2_buf(c->guest_mac, NULL); } switch (ntohs(eh->h_proto)) { @@ -968,7 +997,7 @@ void tap_add_packet(struct ctx *c, ssize_t l2len, char *p) * tap_sock_reset() - Handle closing or failure of connect AF_UNIX socket * @c: Execution context */ -static void tap_sock_reset(struct ctx *c) +void tap_sock_reset(struct ctx *c) { info("Client connection closed%s", c->one_off ? ", exiting" : ""); @@ -979,27 +1008,22 @@ static void tap_sock_reset(struct ctx *c) epoll_ctl(c->epollfd, EPOLL_CTL_DEL, c->fd_tap, NULL); close(c->fd_tap); c->fd_tap = -1; + if (c->mode == MODE_VU) + vu_cleanup(c->vdev); } /** - * tap_handler_passt() - Packet handler for AF_UNIX file descriptor + * tap_passt_input() - Handler for new data on the socket to qemu * @c: Execution context - * @events: epoll events * @now: Current timestamp */ -void tap_handler_passt(struct ctx *c, uint32_t events, - const struct timespec *now) +static void tap_passt_input(struct ctx *c, const struct timespec *now) { static const char *partial_frame; static ssize_t partial_len = 0; ssize_t n; char *p; - if (events & (EPOLLRDHUP | EPOLLHUP | EPOLLERR)) { - tap_sock_reset(c); - return; - } - tap_flush_pools(); if (partial_len) { @@ -1010,10 +1034,13 @@ void tap_handler_passt(struct ctx *c, uint32_t events, memmove(pkt_buf, partial_frame, partial_len); } - n = recv(c->fd_tap, pkt_buf + partial_len, TAP_BUF_BYTES - partial_len, - MSG_DONTWAIT); + do { + n = recv(c->fd_tap, pkt_buf + partial_len, + TAP_BUF_BYTES - partial_len, MSG_DONTWAIT); + } while ((n < 0) && errno == EINTR); + if (n < 0) { - if (errno != EINTR && errno != EAGAIN && errno != EWOULDBLOCK) { + if (errno != EAGAIN && errno != EWOULDBLOCK) { err_perror("Receive error on guest connection, reset"); tap_sock_reset(c); } @@ -1052,54 +1079,76 @@ void tap_handler_passt(struct ctx *c, uint32_t events, } /** - * tap_handler_pasta() - Packet handler for /dev/net/tun file descriptor + * tap_handler_passt() - Event handler for AF_UNIX file descriptor * @c: Execution context * @events: epoll events * @now: Current timestamp */ -void tap_handler_pasta(struct ctx *c, uint32_t events, +void tap_handler_passt(struct ctx *c, uint32_t events, const struct timespec *now) { - ssize_t n, len; - int ret; + if (events & (EPOLLRDHUP | EPOLLHUP | EPOLLERR)) { + tap_sock_reset(c); + return; + } - if (events & (EPOLLRDHUP | EPOLLHUP | EPOLLERR)) - die("Disconnect event on /dev/net/tun device, exiting"); + if (events & EPOLLIN) + tap_passt_input(c, now); +} -redo: - n = 0; +/** + * tap_pasta_input() - Handler for new data on the socket to hypervisor + * @c: Execution context + * @now: Current timestamp + */ +static void tap_pasta_input(struct ctx *c, const struct timespec *now) +{ + ssize_t n, len; tap_flush_pools(); -restart: - while ((len = read(c->fd_tap, pkt_buf + n, TAP_BUF_BYTES - n)) > 0) { - if (len < (ssize_t)sizeof(struct ethhdr) || - len > (ssize_t)ETH_MAX_MTU) { - n += len; - continue; - } + for (n = 0; n <= (ssize_t)(TAP_BUF_BYTES - ETH_MAX_MTU); n += len) { + len = read(c->fd_tap, pkt_buf + n, ETH_MAX_MTU); + if (len == 0) { + die("EOF on tap device, exiting"); + } else if (len < 0) { + if (errno == EINTR) { + len = 0; + continue; + } - tap_add_packet(c, len, pkt_buf + n); + if (errno == EAGAIN && errno == EWOULDBLOCK) + break; /* all done for now */ - if ((n += len) == TAP_BUF_BYTES) - break; - } + die("Error on tap device, exiting"); + } - if (len < 0 && errno == EINTR) - goto restart; + /* Ignore frames of bad length */ + if (len < (ssize_t)sizeof(struct ethhdr) || + len > (ssize_t)ETH_MAX_MTU) + continue; - ret = errno; + tap_add_packet(c, len, pkt_buf + n); + } tap_handler(c, now); +} - if (len > 0 || ret == EAGAIN) - return; - - if (n == TAP_BUF_BYTES) - goto redo; +/** + * tap_handler_pasta() - Packet handler for /dev/net/tun file descriptor + * @c: Execution context + * @events: epoll events + * @now: Current timestamp + */ +void tap_handler_pasta(struct ctx *c, uint32_t events, + const struct timespec *now) +{ + if (events & (EPOLLRDHUP | EPOLLHUP | EPOLLERR)) + die("Disconnect event on /dev/net/tun device, exiting"); - die("Error on tap device, exiting"); + if (events & EPOLLIN) + tap_pasta_input(c, now); } /** @@ -1110,7 +1159,7 @@ restart: */ int tap_sock_unix_open(char *sock_path) { - int fd = socket(AF_UNIX, SOCK_STREAM, 0); + int fd = socket(AF_UNIX, SOCK_STREAM | SOCK_CLOEXEC, 0); struct sockaddr_un addr = { .sun_family = AF_UNIX, }; @@ -1125,10 +1174,12 @@ int tap_sock_unix_open(char *sock_path) if (*sock_path) memcpy(path, sock_path, UNIX_PATH_MAX); - else - snprintf(path, UNIX_PATH_MAX - 1, UNIX_SOCK_PATH, i); + else if (snprintf_check(path, UNIX_PATH_MAX - 1, + UNIX_SOCK_PATH, i)) + die_perror("Can't build UNIX domain socket path"); - ex = socket(AF_UNIX, SOCK_STREAM | SOCK_NONBLOCK, 0); + ex = socket(AF_UNIX, SOCK_STREAM | SOCK_NONBLOCK | SOCK_CLOEXEC, + 0); if (ex < 0) die_perror("Failed to check for UNIX domain conflicts"); @@ -1163,10 +1214,35 @@ int tap_sock_unix_open(char *sock_path) } /** + * tap_backend_show_hints() - Give help information to start QEMU + * @c: Execution context + */ +static void tap_backend_show_hints(struct ctx *c) +{ + switch (c->mode) { + case MODE_PASTA: + /* No hints */ + break; + case MODE_PASST: + info("\nYou can now start qemu (>= 7.2, with commit 13c6be96618c):"); + info(" kvm ... -device virtio-net-pci,netdev=s -netdev stream,id=s,server=off,addr.type=unix,addr.path=%s", + c->sock_path); + info("or qrap, for earlier qemu versions:"); + info(" ./qrap 5 kvm ... -net socket,fd=5 -net nic,model=virtio"); + break; + case MODE_VU: + info("You can start qemu with:"); + info(" kvm ... -chardev socket,id=chr0,path=%s -netdev vhost-user,id=netdev0,chardev=chr0 -device virtio-net,netdev=netdev0 -object memory-backend-memfd,id=memfd0,share=on,size=$RAMSIZE -numa node,memdev=memfd0\n", + c->sock_path); + break; + } +} + +/** * tap_sock_unix_init() - Start listening for connections on AF_UNIX socket * @c: Execution context */ -static void tap_sock_unix_init(struct ctx *c) +static void tap_sock_unix_init(const struct ctx *c) { union epoll_ref ref = { .type = EPOLL_TYPE_TAP_LISTEN }; struct epoll_event ev = { 0 }; @@ -1177,12 +1253,33 @@ static void tap_sock_unix_init(struct ctx *c) ev.events = EPOLLIN | EPOLLET; ev.data.u64 = ref.u64; epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap_listen, &ev); +} + +/** + * tap_start_connection() - start a new connection + * @c: Execution context + */ +static void tap_start_connection(const struct ctx *c) +{ + struct epoll_event ev = { 0 }; + union epoll_ref ref = { 0 }; + + ref.fd = c->fd_tap; + switch (c->mode) { + case MODE_PASST: + ref.type = EPOLL_TYPE_TAP_PASST; + break; + case MODE_PASTA: + ref.type = EPOLL_TYPE_TAP_PASTA; + break; + case MODE_VU: + ref.type = EPOLL_TYPE_VHOST_CMD; + break; + } - info("\nYou can now start qemu (>= 7.2, with commit 13c6be96618c):"); - info(" kvm ... -device virtio-net-pci,netdev=s -netdev stream,id=s,server=off,addr.type=unix,addr.path=%s", - c->sock_path); - info("or qrap, for earlier qemu versions:"); - info(" ./qrap 5 kvm ... -net socket,fd=5 -net nic,model=virtio"); + ev.events = EPOLLIN | EPOLLRDHUP; + ev.data.u64 = ref.u64; + epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap, &ev); } /** @@ -1192,8 +1289,6 @@ static void tap_sock_unix_init(struct ctx *c) */ void tap_listen_handler(struct ctx *c, uint32_t events) { - union epoll_ref ref = { .type = EPOLL_TYPE_TAP_PASST }; - struct epoll_event ev = { 0 }; int v = INT_MAX / 2; struct ucred ucred; socklen_t len; @@ -1232,10 +1327,7 @@ void tap_listen_handler(struct ctx *c, uint32_t events) setsockopt(c->fd_tap, SOL_SOCKET, SO_SNDBUF, &v, sizeof(v))) trace("tap: failed to set SO_SNDBUF to %i", v); - ref.fd = c->fd_tap; - ev.events = EPOLLIN | EPOLLRDHUP; - ev.data.u64 = ref.u64; - epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap, &ev); + tap_start_connection(c); } /** @@ -1261,7 +1353,7 @@ static int tap_ns_tun(void *arg) if (fd < 0) die_perror("Failed to open() /dev/net/tun"); - rc = ioctl(fd, TUNSETIFF, &ifr); + rc = ioctl(fd, (int)TUNSETIFF, &ifr); if (rc < 0) die_perror("TUNSETIFF ioctl on /dev/net/tun failed"); @@ -1279,64 +1371,68 @@ static int tap_ns_tun(void *arg) */ static void tap_sock_tun_init(struct ctx *c) { - union epoll_ref ref = { .type = EPOLL_TYPE_TAP_PASTA }; - struct epoll_event ev = { 0 }; - NS_CALL(tap_ns_tun, c); if (c->fd_tap == -1) die("Failed to set up tap device in namespace"); pasta_ns_conf(c); - ref.fd = c->fd_tap; - ev.events = EPOLLIN | EPOLLRDHUP; - ev.data.u64 = ref.u64; - epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap, &ev); + tap_start_connection(c); } /** - * tap_sock_init() - Create and set up AF_UNIX socket or tuntap file descriptor - * @c: Execution context + * tap_sock_update_pool() - Set the buffer base and size for the pool of packets + * @base: Buffer base + * @size Buffer size */ -void tap_sock_init(struct ctx *c) +void tap_sock_update_pool(void *base, size_t size) { - size_t sz = sizeof(pkt_buf); int i; - pool_tap4_storage = PACKET_INIT(pool_tap4, TAP_MSGS, pkt_buf, sz); - pool_tap6_storage = PACKET_INIT(pool_tap6, TAP_MSGS, pkt_buf, sz); + pool_tap4_storage = PACKET_INIT(pool_tap4, TAP_MSGS, base, size); + pool_tap6_storage = PACKET_INIT(pool_tap6, TAP_MSGS, base, size); for (i = 0; i < TAP_SEQS; i++) { - tap4_l4[i].p = PACKET_INIT(pool_l4, UIO_MAXIOV, pkt_buf, sz); - tap6_l4[i].p = PACKET_INIT(pool_l4, UIO_MAXIOV, pkt_buf, sz); + tap4_l4[i].p = PACKET_INIT(pool_l4, UIO_MAXIOV, base, size); + tap6_l4[i].p = PACKET_INIT(pool_l4, UIO_MAXIOV, base, size); } +} - if (c->fd_tap != -1) { /* Passed as --fd */ - struct epoll_event ev = { 0 }; - union epoll_ref ref; +/** + * tap_backend_init() - Create and set up AF_UNIX socket or + * tuntap file descriptor + * @c: Execution context + */ +void tap_backend_init(struct ctx *c) +{ + if (c->mode == MODE_VU) { + tap_sock_update_pool(NULL, 0); + vu_init(c); + } else { + tap_sock_update_pool(pkt_buf, sizeof(pkt_buf)); + } + if (c->fd_tap != -1) { /* Passed as --fd */ ASSERT(c->one_off); - ref.fd = c->fd_tap; - if (c->mode == MODE_PASST) - ref.type = EPOLL_TYPE_TAP_PASST; - else - ref.type = EPOLL_TYPE_TAP_PASTA; - - ev.events = EPOLLIN | EPOLLRDHUP; - ev.data.u64 = ref.u64; - epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap, &ev); + tap_start_connection(c); return; } - if (c->mode == MODE_PASTA) { + switch (c->mode) { + case MODE_PASTA: tap_sock_tun_init(c); - } else { + break; + case MODE_VU: + case MODE_PASST: tap_sock_unix_init(c); /* In passt mode, we don't know the guest's MAC address until it * sends us packets. Use the broadcast address so that our * first packets will reach it. */ - memset(&c->mac_guest, 0xff, sizeof(c->mac_guest)); + memset(&c->guest_mac, 0xff, sizeof(c->guest_mac)); + break; } + + tap_backend_show_hints(c); } |