summaryrefslogtreecommitdiffstats
path: root/src/libsystemd/sd-netlink/netlink-socket.c
diff options
context:
space:
mode:
Diffstat (limited to 'src/libsystemd/sd-netlink/netlink-socket.c')
-rw-r--r--src/libsystemd/sd-netlink/netlink-socket.c459
1 files changed, 459 insertions, 0 deletions
diff --git a/src/libsystemd/sd-netlink/netlink-socket.c b/src/libsystemd/sd-netlink/netlink-socket.c
new file mode 100644
index 0000000..64cde89
--- /dev/null
+++ b/src/libsystemd/sd-netlink/netlink-socket.c
@@ -0,0 +1,459 @@
+/* SPDX-License-Identifier: LGPL-2.1-or-later */
+
+#include <malloc.h>
+#include <netinet/in.h>
+#include <stdbool.h>
+#include <unistd.h>
+
+#include "sd-netlink.h"
+
+#include "alloc-util.h"
+#include "fd-util.h"
+#include "format-util.h"
+#include "iovec-util.h"
+#include "netlink-internal.h"
+#include "netlink-types.h"
+#include "socket-util.h"
+
+static int broadcast_groups_get(sd_netlink *nl) {
+ _cleanup_free_ uint32_t *groups = NULL;
+ socklen_t len = 0, old_len;
+ int r;
+
+ assert(nl);
+ assert(nl->fd >= 0);
+
+ if (getsockopt(nl->fd, SOL_NETLINK, NETLINK_LIST_MEMBERSHIPS, NULL, &len) < 0) {
+ if (errno != ENOPROTOOPT)
+ return -errno;
+
+ nl->broadcast_group_dont_leave = true;
+ return 0;
+ }
+
+ if (len == 0)
+ return 0;
+
+ groups = new0(uint32_t, len);
+ if (!groups)
+ return -ENOMEM;
+
+ old_len = len;
+
+ if (getsockopt(nl->fd, SOL_NETLINK, NETLINK_LIST_MEMBERSHIPS, groups, &len) < 0)
+ return -errno;
+
+ if (old_len != len)
+ return -EIO;
+
+ for (unsigned i = 0; i < len; i++)
+ for (unsigned j = 0; j < sizeof(uint32_t) * 8; j++)
+ if (groups[i] & (1U << j)) {
+ unsigned group = i * sizeof(uint32_t) * 8 + j + 1;
+
+ r = hashmap_ensure_put(&nl->broadcast_group_refs, NULL, UINT_TO_PTR(group), UINT_TO_PTR(1));
+ if (r < 0)
+ return r;
+ }
+
+ return 0;
+}
+
+int socket_bind(sd_netlink *nl) {
+ socklen_t addrlen;
+ int r;
+
+ r = setsockopt_int(nl->fd, SOL_NETLINK, NETLINK_PKTINFO, true);
+ if (r < 0)
+ return r;
+
+ addrlen = sizeof(nl->sockaddr);
+
+ /* ignore EINVAL to allow binding an already bound socket */
+ if (bind(nl->fd, &nl->sockaddr.sa, addrlen) < 0 && errno != EINVAL)
+ return -errno;
+
+ if (getsockname(nl->fd, &nl->sockaddr.sa, &addrlen) < 0)
+ return -errno;
+
+ return broadcast_groups_get(nl);
+}
+
+static unsigned broadcast_group_get_ref(sd_netlink *nl, unsigned group) {
+ assert(nl);
+
+ return PTR_TO_UINT(hashmap_get(nl->broadcast_group_refs, UINT_TO_PTR(group)));
+}
+
+static int broadcast_group_set_ref(sd_netlink *nl, unsigned group, unsigned n_ref) {
+ int r;
+
+ assert(nl);
+
+ r = hashmap_ensure_allocated(&nl->broadcast_group_refs, NULL);
+ if (r < 0)
+ return r;
+
+ return hashmap_replace(nl->broadcast_group_refs, UINT_TO_PTR(group), UINT_TO_PTR(n_ref));
+}
+
+static int broadcast_group_join(sd_netlink *nl, unsigned group) {
+ assert(nl);
+ assert(nl->fd >= 0);
+ assert(group > 0);
+
+ /* group is "unsigned", but netlink(7) says the argument for NETLINK_ADD_MEMBERSHIP is "int" */
+ return setsockopt_int(nl->fd, SOL_NETLINK, NETLINK_ADD_MEMBERSHIP, group);
+}
+
+int socket_broadcast_group_ref(sd_netlink *nl, unsigned group) {
+ unsigned n_ref;
+ int r;
+
+ assert(nl);
+
+ n_ref = broadcast_group_get_ref(nl, group);
+
+ n_ref++;
+
+ r = broadcast_group_set_ref(nl, group, n_ref);
+ if (r < 0)
+ return r;
+
+ if (n_ref > 1)
+ /* already in the group */
+ return 0;
+
+ return broadcast_group_join(nl, group);
+}
+
+static int broadcast_group_leave(sd_netlink *nl, unsigned group) {
+ assert(nl);
+ assert(nl->fd >= 0);
+ assert(group > 0);
+
+ if (nl->broadcast_group_dont_leave)
+ return 0;
+
+ /* group is "unsigned", but netlink(7) says the argument for NETLINK_DROP_MEMBERSHIP is "int" */
+ return setsockopt_int(nl->fd, SOL_NETLINK, NETLINK_DROP_MEMBERSHIP, group);
+}
+
+int socket_broadcast_group_unref(sd_netlink *nl, unsigned group) {
+ unsigned n_ref;
+ int r;
+
+ assert(nl);
+
+ n_ref = broadcast_group_get_ref(nl, group);
+ if (n_ref == 0)
+ return 0;
+
+ n_ref--;
+
+ r = broadcast_group_set_ref(nl, group, n_ref);
+ if (r < 0)
+ return r;
+
+ if (n_ref > 0)
+ /* still refs left */
+ return 0;
+
+ return broadcast_group_leave(nl, group);
+}
+
+/* returns the number of bytes sent, or a negative error code */
+int socket_write_message(sd_netlink *nl, sd_netlink_message *m) {
+ union sockaddr_union addr = {
+ .nl.nl_family = AF_NETLINK,
+ };
+ ssize_t k;
+
+ assert(nl);
+ assert(m);
+ assert(m->hdr);
+
+ k = sendto(nl->fd, m->hdr, m->hdr->nlmsg_len, 0, &addr.sa, sizeof(addr));
+ if (k < 0)
+ return -errno;
+
+ return k;
+}
+
+static int socket_recv_message(int fd, void *buf, size_t buf_size, uint32_t *ret_mcast_group, bool peek) {
+ struct iovec iov = IOVEC_MAKE(buf, buf_size);
+ union sockaddr_union sender;
+ CMSG_BUFFER_TYPE(CMSG_SPACE(sizeof(struct nl_pktinfo))) control;
+ struct msghdr msg = {
+ .msg_iov = &iov,
+ .msg_iovlen = 1,
+ .msg_name = &sender,
+ .msg_namelen = sizeof(sender),
+ .msg_control = &control,
+ .msg_controllen = sizeof(control),
+ };
+ ssize_t n;
+
+ assert(fd >= 0);
+ assert(peek || (buf && buf_size > 0));
+
+ n = recvmsg_safe(fd, &msg, MSG_TRUNC | (peek ? MSG_PEEK : 0));
+ if (n == -ENOBUFS)
+ return log_debug_errno(n, "sd-netlink: kernel receive buffer overrun");
+ else if (ERRNO_IS_NEG_TRANSIENT(n)) {
+ if (ret_mcast_group)
+ *ret_mcast_group = 0;
+ return 0;
+ } else if (n < 0)
+ return (int) n;
+
+ if (sender.nl.nl_pid != 0) {
+ /* not from the kernel, ignore */
+ log_debug("sd-netlink: ignoring message from PID %"PRIu32, sender.nl.nl_pid);
+
+ if (peek) {
+ /* drop the message */
+ n = recvmsg_safe(fd, &msg, 0);
+ if (n < 0)
+ return (int) n;
+ }
+
+ if (ret_mcast_group)
+ *ret_mcast_group = 0;
+ return 0;
+ }
+
+ if (!peek && (size_t) n > buf_size) /* message did not fit in read buffer */
+ return -EIO;
+
+ if (ret_mcast_group) {
+ struct nl_pktinfo *pi;
+
+ pi = CMSG_FIND_DATA(&msg, SOL_NETLINK, NETLINK_PKTINFO, struct nl_pktinfo);
+ if (pi)
+ *ret_mcast_group = pi->group;
+ else
+ *ret_mcast_group = 0;
+ }
+
+ return (int) n;
+}
+
+DEFINE_PRIVATE_HASH_OPS_WITH_VALUE_DESTRUCTOR(
+ netlink_message_hash_ops,
+ void, trivial_hash_func, trivial_compare_func,
+ sd_netlink_message, sd_netlink_message_unref);
+
+static int netlink_queue_received_message(sd_netlink *nl, sd_netlink_message *m) {
+ uint32_t serial;
+ int r;
+
+ assert(nl);
+ assert(m);
+
+ if (ordered_set_size(nl->rqueue) >= NETLINK_RQUEUE_MAX)
+ return log_debug_errno(SYNTHETIC_ERRNO(ENOBUFS),
+ "sd-netlink: exhausted the read queue size (%d)", NETLINK_RQUEUE_MAX);
+
+ r = ordered_set_ensure_put(&nl->rqueue, &netlink_message_hash_ops, m);
+ if (r < 0)
+ return r;
+
+ sd_netlink_message_ref(m);
+
+ if (sd_netlink_message_is_broadcast(m))
+ return 0;
+
+ serial = message_get_serial(m);
+ if (serial == 0)
+ return 0;
+
+ if (sd_netlink_message_get_errno(m) < 0) {
+ _cleanup_(sd_netlink_message_unrefp) sd_netlink_message *old = NULL;
+
+ old = hashmap_remove(nl->rqueue_by_serial, UINT32_TO_PTR(serial));
+ if (old)
+ log_debug("sd-netlink: received error message with serial %"PRIu32", but another message with "
+ "the same serial is already stored in the read queue, replacing.", serial);
+ }
+
+ r = hashmap_ensure_put(&nl->rqueue_by_serial, &netlink_message_hash_ops, UINT32_TO_PTR(serial), m);
+ if (r == -EEXIST) {
+ if (!sd_netlink_message_is_error(m))
+ log_debug("sd-netlink: received message with serial %"PRIu32", but another message with "
+ "the same serial is already stored in the read queue, ignoring.", serial);
+ return 0;
+ }
+ if (r < 0) {
+ sd_netlink_message_unref(ordered_set_remove(nl->rqueue, m));
+ return r;
+ }
+
+ sd_netlink_message_ref(m);
+ return 0;
+}
+
+static int netlink_queue_partially_received_message(sd_netlink *nl, sd_netlink_message *m) {
+ uint32_t serial;
+ int r;
+
+ assert(nl);
+ assert(m);
+ assert(m->hdr->nlmsg_flags & NLM_F_MULTI);
+
+ if (hashmap_size(nl->rqueue_partial_by_serial) >= NETLINK_RQUEUE_MAX)
+ return log_debug_errno(SYNTHETIC_ERRNO(ENOBUFS),
+ "sd-netlink: exhausted the partial read queue size (%d)", NETLINK_RQUEUE_MAX);
+
+ serial = message_get_serial(m);
+ r = hashmap_ensure_put(&nl->rqueue_partial_by_serial, &netlink_message_hash_ops, UINT32_TO_PTR(serial), m);
+ if (r < 0)
+ return r;
+
+ sd_netlink_message_ref(m);
+ return 0;
+}
+
+static int parse_message_one(sd_netlink *nl, uint32_t group, const struct nlmsghdr *hdr, sd_netlink_message **ret) {
+ _cleanup_(sd_netlink_message_unrefp) sd_netlink_message *m = NULL;
+ size_t size;
+ int r;
+
+ assert(nl);
+ assert(hdr);
+ assert(ret);
+
+ /* not broadcast and not for us */
+ if (group == 0 && hdr->nlmsg_pid != nl->sockaddr.nl.nl_pid)
+ goto finalize;
+
+ /* silently drop noop messages */
+ if (hdr->nlmsg_type == NLMSG_NOOP)
+ goto finalize;
+
+ /* check that we support this message type */
+ r = netlink_get_policy_set_and_header_size(nl, hdr->nlmsg_type, NULL, &size);
+ if (r == -EOPNOTSUPP) {
+ log_debug("sd-netlink: ignored message with unknown type: %i", hdr->nlmsg_type);
+ goto finalize;
+ }
+ if (r < 0)
+ return r;
+
+ /* check that the size matches the message type */
+ if (hdr->nlmsg_len < NLMSG_LENGTH(size)) {
+ log_debug("sd-netlink: message is shorter than expected, dropping.");
+ goto finalize;
+ }
+
+ r = message_new_empty(nl, &m);
+ if (r < 0)
+ return r;
+
+ m->multicast_group = group;
+ m->hdr = memdup(hdr, hdr->nlmsg_len);
+ if (!m->hdr)
+ return -ENOMEM;
+
+ /* seal and parse the top-level message */
+ r = sd_netlink_message_rewind(m, nl);
+ if (r < 0)
+ return r;
+
+ *ret = TAKE_PTR(m);
+ return 1;
+
+finalize:
+ *ret = NULL;
+ return 0;
+}
+
+/* On success, the number of bytes received is returned and *ret points to the received message
+ * which has a valid header and the correct size.
+ * If nothing useful was received 0 is returned.
+ * On failure, a negative error code is returned.
+ */
+int socket_read_message(sd_netlink *nl) {
+ bool done = false;
+ uint32_t group;
+ size_t len;
+ int r;
+
+ assert(nl);
+
+ /* read nothing, just get the pending message size */
+ r = socket_recv_message(nl->fd, NULL, 0, NULL, true);
+ if (r <= 0)
+ return r;
+ len = (size_t) r;
+
+ /* make room for the pending message */
+ if (!greedy_realloc((void**) &nl->rbuffer, len, sizeof(uint8_t)))
+ return -ENOMEM;
+
+ /* read the pending message */
+ r = socket_recv_message(nl->fd, nl->rbuffer, MALLOC_SIZEOF_SAFE(nl->rbuffer), &group, false);
+ if (r <= 0)
+ return r;
+ len = (size_t) r;
+
+ if (!NLMSG_OK(nl->rbuffer, len)) {
+ log_debug("sd-netlink: received invalid message, discarding %zu bytes of incoming message", len);
+ return 0;
+ }
+
+ for (struct nlmsghdr *hdr = nl->rbuffer; NLMSG_OK(hdr, len); hdr = NLMSG_NEXT(hdr, len)) {
+ _cleanup_(sd_netlink_message_unrefp) sd_netlink_message *m = NULL;
+
+ r = parse_message_one(nl, group, hdr, &m);
+ if (r < 0)
+ return r;
+ if (r == 0)
+ continue;
+
+ if (hdr->nlmsg_flags & NLM_F_MULTI) {
+ if (hdr->nlmsg_type == NLMSG_DONE) {
+ _cleanup_(sd_netlink_message_unrefp) sd_netlink_message *existing = NULL;
+
+ /* finished reading multi-part message */
+ existing = hashmap_remove(nl->rqueue_partial_by_serial, UINT32_TO_PTR(hdr->nlmsg_seq));
+
+ /* if we receive only NLMSG_DONE, put it into the receive queue. */
+ r = netlink_queue_received_message(nl, existing ?: m);
+ if (r < 0)
+ return r;
+
+ done = true;
+ } else {
+ sd_netlink_message *existing;
+
+ existing = hashmap_get(nl->rqueue_partial_by_serial, UINT32_TO_PTR(hdr->nlmsg_seq));
+ if (existing) {
+ /* This is the continuation of the previously read messages.
+ * Let's append this message at the end. */
+ while (existing->next)
+ existing = existing->next;
+ existing->next = TAKE_PTR(m);
+ } else {
+ /* This is the first message. Put it into the queue for partially
+ * received messages. */
+ r = netlink_queue_partially_received_message(nl, m);
+ if (r < 0)
+ return r;
+ }
+ }
+
+ } else {
+ r = netlink_queue_received_message(nl, m);
+ if (r < 0)
+ return r;
+
+ done = true;
+ }
+ }
+
+ if (len > 0)
+ log_debug("sd-netlink: discarding trailing %zu bytes of incoming message", len);
+
+ return done;
+}
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-20 00:13:15 +0000