From 5e9d03d507aad324a803eb8795e0eed6fb671761 Mon Sep 17 00:00:00 2001
From: guowenxue <guowenxue@gmail.com>
Date: Mon, 10 Jul 2023 15:24:52 +0800
Subject: [PATCH] Merge branch 'master' of http://master.iot-yun.club:8088/r/raspberrypi

---
 apue/project_socket/src/socket.c |  515 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 515 insertions(+), 0 deletions(-)

diff --git a/apue/project_socket/src/socket.c b/apue/project_socket/src/socket.c
new file mode 100644
index 0000000..e6231e3
--- /dev/null
+++ b/apue/project_socket/src/socket.c
@@ -0,0 +1,515 @@
+/********************************************************************************
+ *      Copyright:  (C) 2022 LingYun IoT System Studio
+ *                  All rights reserved.
+ *
+ *       Filename:  socket.c
+ *    Description:  This file is for socket API functions
+ *
+ *        Version:  1.0.0(18/04/22)
+ *         Author:  Guo Wenxue <guowenxue@gmail.com>
+ *      ChangeLog:  1, Release initial version on "18/04/22 17:09:59"
+ *
+ ********************************************************************************/
+#include <stdio.h>
+#include <unistd.h>
+#include <string.h>
+#include <netdb.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <sys/un.h>
+#include <poll.h>
+#include <errno.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <linux/sockios.h>
+#include <sys/ioctl.h>
+#include <sys/socket.h>
+#include <netinet/in.h>
+#include <netinet/tcp.h>
+#include <arpa/inet.h>
+#include <sys/resource.h>
+
+#include "socket.h"
+#include "logger.h"
+
+/*  description: initial socket context
+ *   input args:
+ *               $sock:  socket context pointer
+ *               $host:  connect server hostname for client mode, unused for server mode
+ *               $port:  connect server port for client mode or listen port for server mode
+ * return value: <0: failure   0:ok
+ */
+int socket_init(socket_ctx_t *sock, char *host, int port)
+{
+	if( !sock || port<=0 )
+		return -1;
+
+	memset( sock, 0, sizeof(*sock) );
+	sock->fd = -1;
+	sock->port = port;
+    if( host ) /* server no need it */
+        strncpy(sock->host, host, HOSTNAME_LEN);
+}
+
+/*  description: close socket
+ *   input args:
+ *               $sock:  socket context pointer
+ * return value: <0: failure   0:ok
+ */
+int socket_term(socket_ctx_t *sock)
+{
+	if( !sock )
+		return -1;
+
+	if( sock->fd > 0)
+	{
+		close(sock->fd);
+		sock->fd = -1;
+	}
+
+	return 0;
+}
+
+/*  description: socket server start listen
+ *   input args:
+ *               $sock:  socket context pointer
+ * return value: <0: failure   0:ok
+ */
+int socket_listen(socket_ctx_t *sock)
+{
+    int                 rv = 0;
+    struct sockaddr_in  addr;
+    int                 backlog = 13;
+
+	if( !sock )
+		return -1;
+
+    set_socket_rlimit(); /* set max open socket count */
+}
+
+/*  description: socket connect to server in block mode
+ *   input args:
+ *               $sock:  socket context pointer
+ * return value: <0: failure   0:ok
+ */
+int socket_connect(socket_ctx_t *sock)
+{
+    int                 rv = 0;
+    int                 sockfd = 0;
+	struct sockaddr_in  servaddr;
+    char                service[20];
+    struct addrinfo     hints, *rp;
+    struct addrinfo    *res = NULL;
+    struct in_addr      inaddr;
+    struct sockaddr_in  addr;
+    int                 len = sizeof(addr);
+
+	if( !sock )
+		return -1;
+
+	socket_term(sock);
+
+    /*+--------------------------------------------------+
+     *| use getaddrinfo() to do domain name translation  |
+     *+--------------------------------------------------+*/
+
+    memset(&hints, 0, sizeof(struct addrinfo));
+    hints.ai_family = AF_INET; /* Only support IPv4 */
+    hints.ai_socktype = SOCK_STREAM;
+    hints.ai_protocol = IPPROTO_TCP; /* TCP protocol */
+
+    /* If $host is a valid IP address, then don't use name resolution */
+    if( inet_aton(sock->host, &inaddr) )
+    {
+        //log_info("%s is a valid IP address, don't use domain name resolution.\n", sock->host);
+        hints.ai_flags |= AI_NUMERICHOST;
+    }
+
+    /* Obtain address(es) matching host/port */
+    snprintf(service, sizeof(service), "%d", sock->port);
+    if( (rv=getaddrinfo(sock->host, service, &hints, &res)) )
+    {
+        log_error("getaddrinfo() parser [%s:%s] failed: %s\n", sock->host, service, gai_strerror(rv));
+        return -3;
+    }
+
+    /* getaddrinfo() returns a list of address structures. Try each
+       address until we successfully connect or bind */
+    for (rp=res; rp!=NULL; rp=rp->ai_next)
+    {
+#if 0
+        char                  ipaddr[INET_ADDRSTRLEN];
+        struct sockaddr_in   *sp = (struct sockaddr_in *) rp->ai_addr;
+
+        /* print domain name translation result */
+        memset( ipaddr, 0, sizeof(ipaddr) );
+        if( inet_ntop(AF_INET, &sp->sin_addr, ipaddr, sizeof(ipaddr)) )
+        {
+            log_info("domain name resolution [%s->%s]\n", sock->host, ipaddr);
+        }
+#endif
+
+        /*  Create the socket */
+        sockfd = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);
+        if( sockfd < 0)
+        {
+            log_error("socket() create failed: %s\n", strerror(errno));
+            rv = -3;
+            continue;
+        }
+
+        /* connect to server */
+        rv = connect(sockfd, rp->ai_addr, len);
+        if( 0 == rv )
+        {
+            sock->fd = sockfd;
+            log_info("Connect to server[%s:%d] on fd[%d] successfully!\n", sock->host, sock->port, sockfd);
+            break;
+        }
+        else
+        {
+            /* socket connect get error, try another IP address */
+            close(sockfd);
+            continue;
+        }
+    }
+
+    freeaddrinfo(res);
+    return rv;
+}
+
+/*  description: send data from the socket
+ *   input args:
+ *               $sock :  socket context pointer
+ *               $data :  socket send data
+ *               $bytes:  socket send data bytes
+ * return value: <0: failure   0:ok
+ */
+int socket_send(socket_ctx_t *sock, char *data, int bytes)
+{
+    int            rv = 0;
+    int            i = 0;
+    int            left_bytes = bytes;
+
+	if( !sock || !data || bytes<= 0 )
+		return -1;
+
+    while( left_bytes > 0 )
+    {
+        rv=write(sock->fd, &data[i], left_bytes);
+        if( rv < 0 )
+        {
+            log_info("socket[%d] write() failure: %s, close socket now\n", sock->fd, strerror(errno));
+            socket_term(sock);
+            return -2;
+        }
+        else if( rv == left_bytes )
+        {
+            log_info("socket send %d bytes data over\n", bytes);
+            return 0;
+        }
+        else
+        {
+            /* not send over this time, continue to send left data  */
+            i += rv;
+            left_bytes -= rv;
+            continue;
+        }
+    }
+}
+
+/*  description: receive data from the socket
+ *   input args:
+ *               $sock :  socket context pointer
+ *               $buf  :  socket receive data buffer
+ *               $size :  socket receive data buffer size
+ *               $timeout: receive data time, <=0 will don't timeout
+ * return value: <0: failure   0:ok
+ */
+int socket_recv(socket_ctx_t *sock, char *buf, int size, int timeout)
+{
+    int               rv = 0;
+    int               i = 0;
+    fd_set            rdset;
+    int               maxfd;
+
+	if( !sock || !buf || size<= 0 )
+		return -1;
+
+    memset(buf, 0, size);
+
+    maxfd = sock->fd;
+    FD_ZERO(&rdset);
+    FD_SET(sock->fd, &rdset);
+
+    if( timeout <= 0 ) /* no timeout  */
+    {
+        rv=select(maxfd+1, &rdset, NULL, NULL, NULL);
+    }
+    else
+    {
+        struct timeval    tv;
+        tv.tv_sec = timeout;
+        tv.tv_usec = 0;
+        rv=select(maxfd+1, &rdset, NULL, NULL, &tv);
+    }
+
+    if( rv < 0 )
+    {
+        log_error("select() on socket[%d] got error: %s\n", sock->fd, strerror(errno));
+        return -2;
+    }
+    else if( rv == 0 )
+    {
+        log_error("select() on socket[%d] get timeout\n", sock->fd);
+        return 0;
+    }
+    else
+    {
+        rv = read(sock->fd, buf, size);
+        if( rv <= 0 )
+        {
+            log_error("socket[%d] read() failure or got disconnected: %s, close socket now\n", sock->fd, strerror(errno));
+            socket_term(sock);
+            return -2;
+        }
+        else
+        {
+            log_debug("socket[%d] receive %d bytes data\n", sock->fd, rv);
+            return rv;
+        }
+    }
+}
+
+
+
+/*+-------------------------------------------------------------------+
+ *|                socket utils function                              |
+ *+-------------------------------------------------------------------+*/
+
+
+/*  socket connected or not: <0: failure  0:ok */
+int sock_check_connect(int sockfd)
+{
+	struct tcp_info   info;
+	int               len=sizeof(info);
+
+	if( sockfd < 0 )
+		return -1;
+
+	getsockopt(sockfd, IPPROTO_TCP, TCP_INFO, &info, (socklen_t *)&len);
+
+	if( TCP_CLOSE==info.tcpi_state || TCP_CLOSING==info.tcpi_state || TCP_CLOSE_WAIT==info.tcpi_state )
+ 	{
+		return -3;
+	}
+
+	return -0;
+}
+
+/* description: set socket listen port as reusable, fix port already used bug  */
+int socket_set_reuseaddr(int sockfd)
+{
+    int opt = 1;
+    int len = sizeof (int);
+
+    if (setsockopt (sockfd, SOL_SOCKET, SO_REUSEADDR, (void *) &opt, len))
+    {
+        log_error("Set socket[%d] option SO_REUSEADDR failed:%s\n", sockfd, strerror(errno));
+        return -1;
+    }
+    log_debug("Set socket[%d] option SO_REUSEADDR ok\n", sockfd);
+
+    return 0;
+}
+
+/* set socket as non-block mode, common socket default work as block mode */
+int socket_set_nonblock(int sockfd)
+{
+    int opts;
+    /*
+     * fcntl may set:
+     *
+     * EACCES, EAGAIN: Operation is prohibited by locks held by other
+     *          processes. Or, operation is prohibited because the file has
+     *          been memory-mapped by another process.
+     * EBADF:   fd is not an open file descriptor, or the command was F_SETLK
+     *          or F_SETLKW and the file descriptor open mode doesn't match
+     *          with the type of lock requested.
+     * EDEADLK: It was detected that the specified F_SETLKW command would
+     *          cause a deadlock.
+     * EFAULT:  lock is outside your accessible address space.
+     * EINTR:   For F_SETLKW, the command was interrupted by a signal. For
+     *          F_GETLK and F_SETLK, the command was interrupted by a signal
+     *          before the lock was checked or acquired. Most likely when
+     *          locking a remote file (e.g. locking over NFS), but can
+     *          sometimes happen locally.
+     * EINVAL:  For F_DUPFD, arg is negative or is greater than the maximum
+     *          allowable value. For F_SETSIG, arg is not an allowable signal
+     *          number.
+     * EMFILE:  For F_DUPFD, the process already has the maximum number of
+     *          file descriptors open.
+     * ENOLCK:  Too many segment locks open, lock table is full, or a remote
+     *          locking protocol failed (e.g. locking over NFS).
+     * EPERM:   Attempted to clear the O_APPEND flag on a file that has the
+     *          append-only attribute set.
+     */
+    opts = fcntl(sockfd, F_GETFL);
+    if (opts < 0)
+    {
+        log_warn("fcntl() get socket options failure: %s\n", strerror(errno));
+        return -1;
+    }
+
+    opts |= O_NONBLOCK;
+
+    if (fcntl(sockfd, F_SETFL, opts) < 0)
+    {
+        log_warn("fcntl() set socket options failure: %s\n", strerror(errno));
+        return -1;
+    }
+
+    log_debug("Set socket[%d] none blocking\n", sockfd);
+    return opts;
+}
+
+
+/* set socket receive and send buffer size in linux kernel space */
+int socket_set_buffer(int sockfd, int rsize, int ssize)
+{
+    int        opt;
+    socklen_t  optlen = sizeof(opt);
+
+    if(sockfd < 0)
+        return -1;
+
+    /* Get system default receive buffer size, Linux X86: 85K */
+    if (getsockopt (sockfd, SOL_SOCKET, SO_RCVBUF, (char *) &opt, &optlen))
+    {
+        log_warn("getsockopt() get receive buffer failure: %s\n", strerror(errno));
+        return -2;
+    }
+
+    /* Only when current receive buffer size larger than the default one will change it  */
+    if(rsize > opt)
+    {
+        opt = (int) rsize;
+        if (setsockopt (sockfd, SOL_SOCKET, SO_RCVBUF, (char *) &opt, optlen))
+        {
+            log_warn("setsockopt() set receive buffer to %d failure: %s\n", opt, strerror(errno));
+            return -2;
+        }
+    }
+
+    /* Get system default send buffer size, Linux X86: 16K */
+    if (getsockopt (sockfd, SOL_SOCKET, SO_SNDBUF, (char *) &opt, &optlen))
+    {
+        log_warn("getsockopt() get send buffer failure: %s\n", strerror(errno));
+        return -3;
+    }
+
+    /* Only when current receive buffer size larger than the default one will change it  */
+    if(ssize > opt)
+    {
+        opt = (int) ssize;
+        if (setsockopt (sockfd, SOL_SOCKET, SO_SNDBUF, (char *) &opt, optlen))
+        {
+            log_warn("setsockopt() set send buffer to %d failure: %s\n", opt, strerror(errno));
+            return -3;
+        }
+    }
+
+    log_info("Set socket[%d] RCVBUF size:%d  SNDBUF size:%d\n", sockfd, rsize, ssize);
+    return 0;
+}
+
+/*
+ * Enable socket SO_KEEPALIVE, if the connection disconnected, any system call on socket
+ * will return immediately and errno will be set to "WSAENOTCONN"
+ *
+ * keepalive is not program related, but socket related, * so if you have multiple sockets,
+ * you can handle keepalive for each of them separately.
+ *
+ * Reference: http://tldp.org/HOWTO/html_single/TCP-Keepalive-HOWTO/
+ */
+int socket_set_keepalive(int sockfd, int keepintvl, int keepcnt)
+{
+    int  opt;
+
+    if(sockfd < 0)
+        return -1;
+
+    /* Enable the KEEPALIVE flag */
+    opt = 1;
+    if (setsockopt (sockfd, SOL_SOCKET, SO_KEEPALIVE, (char *) &opt, sizeof (opt)))
+    {
+        log_warn("setsockopt() enable SO_KEEPALIVE failure: %s\n", strerror(errno));
+        return -2;
+    }
+
+    if(keepintvl || keepcnt)
+    {
+        /*
+         *  The tcp_keepidle parameter specifies the interval between the last data packet sent
+         *  (simple ACKs are not considered data) and the first keepalive probe; after the
+         *  connection is marked to need keepalive, this counter is not used any further.
+         *  ~ >: cat /proc/sys/net/ipv4/tcp_keepalive_time
+         *  7200
+         */
+        opt = 3; /* 3 seconds  */
+        if (setsockopt (sockfd, SOL_TCP, TCP_KEEPIDLE, (char *) &opt, sizeof (opt)))
+        {
+            log_error("setsockopt() set TCP_KEEPIDLE to %d seconds failure: %s\n", opt, strerror(errno));
+            return -3;
+        }
+
+        if((opt=keepintvl) > 0)
+        {
+            /*
+             * The tcp_keepintvl parameter specifies the interval between subsequential keepalive
+             * probes, regardless of what the connection has exchanged in the meantime.
+             * ~ >: cat /proc/sys/net/ipv4/tcp_keepalive_intvl
+             * 75
+             */
+            if (setsockopt (sockfd, SOL_TCP, TCP_KEEPINTVL, (char *) &opt, sizeof (opt)))
+            {
+                log_error("setsockopt() set TCP_KEEPINTVL to %d failure: %s\n", opt, strerror(errno));
+                return -4;
+            }
+        }
+
+        if((opt=keepcnt) > 0)
+        {
+            /*
+             * The TCP_KEEPCNT option specifies the maximum number of unacknowledged probes to
+             * send before considering the connection dead and notifying the application layer
+             * probes to be sent. The value of TCP_KEEPCNT is an integer value between 1 and n,
+             * where n is the value of the systemwide tcp_keepcnt parameter.
+             * ~ >: cat /proc/sys/net/ipv4/tcp_keepalive_probes
+             * 9
+             */
+            if (setsockopt (sockfd, SOL_TCP, TCP_KEEPCNT, (char *) &opt, sizeof (opt)))
+            {
+                log_error("setsockopt() set TCP_KEEPCNT to %d failure: %s\n", opt, strerror(errno));
+                return -5;
+            }
+        }
+    }
+
+    log_debug("Set socket[%d] KEEPINTVL:%d  KEEPCNT:%d\n", sockfd, keepintvl, keepcnt);
+    return 0;
+}
+
+
+/* Set open file description count to max */
+void set_socket_rlimit(void)
+{
+    struct rlimit limit = {0};
+
+    getrlimit(RLIMIT_NOFILE, &limit );
+    limit.rlim_cur  = limit.rlim_max;
+    setrlimit(RLIMIT_NOFILE, &limit );
+
+    log_info("set socket open fd max count to %d\n", limit.rlim_max);
+}
+

--
Gitblit v1.9.1