kern: Implement skeleton for POSIX sockets

- Adds POSIX socket()
- Adds POSIX listen()
- Adds vnode socket type
- Adds tmpfs socket awareness

Much work is to be done but this commit provides a solid skeleton
to be built upon.

Signed-off-by: Ian Moffett <ian@osmora.org>

1 files changed, 256 insertions, 0 deletions

diff --git a/src/sys/os/os_socket.c b/src/sys/os/os_socket.c
new file mode 100644
index 0000000..a37d15f
--- /dev/null
+++ b/src/sys/os/os_socket.c
@@ -0,0 +1,256 @@
+/*
+ * Copyright (c) 2025 Ian Marco Moffett and L5 engineers
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright notice,
+ *    this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the project nor the names of its
+ *    contributors may be used to endorse or promote products derived from
+ *    this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <sys/socket.h>
+#include <sys/mount.h>
+#include <sys/syslog.h>
+#include <sys/errno.h>
+#include <sys/proc.h>
+#include <os/vnode.h>
+#include <os/kalloc.h>
+#include <os/filedesc.h>
+#include <string.h>
+
+static size_t next_sockid = 1;
+
+/*
+ * Allocate a socket file descriptor
+ *
+ * @fdp_res: Result is written here
+ *
+ * XXX: Socket vnodes are never deallocated until explicitly
+ *      closed
+ *
+ * Returns file descriptor on success, otherwise a
+ * less than zero value on faulure
+ */
+static int
+get_sock_fd(struct filedesc **fdp_res)
+{
+    char namebuf[256];
+    struct ksocket *sock;
+    struct mount *mp;
+    struct nameidata nd;
+    struct filedesc *fdp;
+    struct proc *self = proc_self();
+    struct vnode *vp;
+    int error;
+
+    if (fdp_res == NULL) {
+        return -EINVAL;
+    }
+
+    if (self == NULL) {
+        return -EIO;
+    }
+
+    error = mount_lookup("/tmp", &mp);
+    if (error < 0) {
+        printf("socket: could not get /tmp mountpoint\n");
+        return error;
+    }
+
+    error = fd_alloc(self, &fdp);
+    if (error < 0) {
+        printf("socket: could not allocate fd\n");
+        return error;
+    }
+
+    sock = kalloc(sizeof(*sock));
+    if (sock == NULL) {
+        fd_close(fdp->fdno);
+        return -EINVAL;
+    }
+
+    /* Format the name */
+    snprintf(
+        namebuf,
+        sizeof(namebuf),
+        "sock.%d",
+        next_sockid++
+    );
+
+    /* Allocate a persistent vnode */
+    error = vfs_valloc(&vp, VTYPE_SOCK, 0);
+    if (error < 0) {
+        printf("socket: could not allocate vnode\n");
+        fd_close(fdp->fdno);
+        kfree(sock);
+        return error;
+    }
+
+    /* Create the backing file */
+    nd.path = namebuf;
+    error = vop_create(mp->vp, &nd, VTYPE_SOCK);
+    if (error < 0) {
+        printf("socket: failed to create /tmp entry\n");
+        fd_close(fdp->fdno);
+        kfree(sock);
+        return error;
+    }
+
+    /* Mark the socket as inactive */
+    memset(sock, 0, sizeof(*sock));
+    sock->backlog = -1;
+
+    fdp->vp = vp;
+    vp->data = sock;
+    printf("socket: socket created @ /tmp/%s\n", namebuf);
+    return fdp->fdno;
+}
+
+/*
+ * Get a socket descriptor from a fd
+ *
+ * @fd: File descriptor to look up
+ * @sock_res: Result pointer is written here
+ *
+ * Returns zero on success, otherwise a less
+ * than zero value on failure
+ */
+static int
+get_sock(int fd, struct ksocket **sock_res)
+{
+    struct filedesc *fdp;
+    struct proc *self = proc_self();
+    struct vnode *vp;
+
+    if (fd < 0) {
+        return -EBADF;
+    }
+
+    if (sock_res == NULL) {
+        return -1;
+    }
+
+    if (self == NULL) {
+        return -EINVAL;
+    }
+
+    /* Get the file descriptor */
+    fdp = fd_get(
+        self,
+        fd
+    );
+
+    if (fdp == NULL) {
+        return -EBADF;
+    }
+
+    /* We need the vnode */
+    if ((vp = fdp->vp) == NULL) {
+        return -EIO;
+    }
+
+    *sock_res = vp->data;
+    return 0;
+}
+
+/*
+ * Any backlog value > 0 marks the socket
+ * as accepting connections
+ */
+int
+listen(int socket, int backlog)
+{
+    struct ksocket *ksock = NULL;
+    int error;
+
+    if (socket < 0) {
+        return -EBADF;
+    }
+
+    if (backlog < 0) {
+        backlog = 0;
+    }
+
+    error = get_sock(socket, &ksock);
+    if (error < 0) {
+        printf("listen: failed to get socket descriptor\n");
+        return error;
+    }
+
+    /* Set the backlog */
+    ksock->backlog = backlog;
+    return 0;
+}
+
+/*
+ * Handles the AF_UNIX domain
+ */
+static int
+af_unix(int type, int protocol)
+{
+    struct filedesc *desc;
+
+    return get_sock_fd(&desc);
+}
+
+int
+socket(int domain, int type, int protocol)
+{
+    if (domain < 0 || type < 0) {
+        return -EINVAL;
+    }
+
+    switch (domain) {
+    case AF_UNIX:
+        return af_unix(type, protocol);
+    }
+
+    return -1;
+}
+
+/*
+ * ARG0: Domain
+ * ARG1: Type
+ * ARG2: Protocol
+ */
+scret_t
+sys_socket(struct syscall_args *scargs)
+{
+    int domain = SCARG(scargs, int, 0);
+    int type = SCARG(scargs, int, 1);
+    int protocol = SCARG(scargs, int, 2);
+
+    return socket(domain, type, protocol);
+}
+
+/*
+ * ARG0: Socket fd
+ * @backlog: Maximum connections
+ */
+scret_t
+sys_listen(struct syscall_args *scargs)
+{
+    int sockfd = SCARG(scargs, int, 0);
+    int backlog = SCARG(scargs, int, 1);
+
+    return listen(sockfd, backlog);
+}