socket-util: Use correct address family in set_dscp(), instead of guessing.

[cascardo/ovs.git] / lib / socket-util.c

diff --git a/lib/socket-util.c b/lib/socket-util.c
index a13a59c..e32aa2b 100644 (file)
--- a/lib/socket-util.c
+++ b/lib/socket-util.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2008, 2009, 2010, 2011, 2012, 2013, 2014 Nicira, Inc.
+ * Copyright (c) 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015 Nicira, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -27,7 +27,6 @@
 #include <stdlib.h>
 #include <string.h>
 #include <sys/ioctl.h>
-#include <sys/resource.h>
 #include <sys/socket.h>
 #include <sys/stat.h>
 #include <sys/uio.h>
@@ -40,7 +39,7 @@
 #include "poll-loop.h"
 #include "util.h"
 #include "vlog.h"
-#if AF_PACKET && LINUX_DATAPATH
+#ifdef __linux__
 #include <linux/if_packet.h>
 #endif
 #ifdef HAVE_NETLINK
@@ -52,9 +51,11 @@ VLOG_DEFINE_THIS_MODULE(socket_util);
 
 /* #ifdefs make it a pain to maintain code: you have to try to build both ways.
  * Thus, this file compiles all of the code regardless of the target, by
- * writing "if (LINUX_DATAPATH)" instead of "#ifdef __linux__". */
-#ifndef LINUX_DATAPATH
-#define LINUX_DATAPATH 0
+ * writing "if (LINUX)" instead of "#ifdef __linux__". */
+#ifdef __linux__
+#define LINUX 1
+#else
+#define LINUX 0
 #endif
 
 #ifndef O_DIRECTORY
@@ -105,21 +106,34 @@ xset_nonblocking(int fd)
     }
 }
 
+/* Sets the DSCP value of socket 'fd' to 'dscp', which must be 63 or less.
+ * 'family' must indicate the socket's address family (AF_INET or AF_INET6, to
+ * do anything useful). */
 int
-set_dscp(int fd, uint8_t dscp)
+set_dscp(int fd, int family, uint8_t dscp)
 {
+    int retval;
     int val;
 
     if (dscp > 63) {
         return EINVAL;
     }
-
     val = dscp << 2;
-    if (setsockopt(fd, IPPROTO_IP, IP_TOS, &val, sizeof val)) {
-        return sock_errno();
+
+    switch (family) {
+    case AF_INET:
+        retval = setsockopt(fd, IPPROTO_IP, IP_TOS, &val, sizeof val);
+        break;
+
+    case AF_INET6:
+        retval = setsockopt(fd, IPPROTO_IPV6, IPV6_TCLASS, &val, sizeof val);
+        break;
+
+    default:
+        return ENOPROTOOPT;
     }
 
-    return 0;
+    return retval ? sock_errno() : 0;
 }
 
 /* Translates 'host_name', which must be a string representation of an IP
@@ -251,6 +265,7 @@ check_connection_completion(int fd)
     }
 }
 
+#ifndef _WIN32
 /* Drain all the data currently in the receive queue of a datagram socket (and
  * possibly additional data).  There is no way to know how many packets are in
  * the receive queue, but we do know that the total number of bytes queued does
@@ -275,8 +290,9 @@ drain_rcvbuf(int fd)
          *
          * On other Unix-like OSes, MSG_TRUNC has no effect in the flags
          * argument. */
-        char buffer[LINUX_DATAPATH ? 1 : 2048];
-        ssize_t n_bytes = recv(fd, buffer, sizeof buffer, MSG_TRUNC);
+        char buffer[LINUX ? 1 : 2048];
+        ssize_t n_bytes = recv(fd, buffer, sizeof buffer,
+                               MSG_TRUNC | MSG_DONTWAIT);
         if (n_bytes <= 0 || n_bytes >= rcvbuf) {
             break;
         }
@@ -284,6 +300,7 @@ drain_rcvbuf(int fd)
     }
     return 0;
 }
+#endif
 
 /* Returns the size of socket 'sock''s receive buffer (SO_RCVBUF), or a
  * negative errno value if an error occurs. */
@@ -331,7 +348,7 @@ shorten_name_via_proc(const char *name, char short_name[MAX_UN_LEN + 1],
     int dirfd;
     int len;
 
-    if (!LINUX_DATAPATH) {
+    if (!LINUX) {
         return ENAMETOOLONG;
     }
 
@@ -749,9 +766,9 @@ inet_open_active(int style, const char *target, uint16_t default_port,
     /* The dscp bits must be configured before connect() to ensure that the
      * TOS field is set during the connection establishment.  If set after
      * connect(), the handshake SYN frames will be sent with a TOS of 0. */
-    error = set_dscp(fd, dscp);
+    error = set_dscp(fd, ss.ss_family, dscp);
     if (error) {
-        VLOG_ERR("%s: socket: %s", target, sock_strerror(error));
+        VLOG_ERR("%s: set_dscp: %s", target, sock_strerror(error));
         goto exit;
     }
 
@@ -886,9 +903,9 @@ inet_open_passive(int style, const char *target, int default_port,
     /* The dscp bits must be configured before connect() to ensure that the TOS
      * field is set during the connection establishment.  If set after
      * connect(), the handshake SYN frames will be sent with a TOS of 0. */
-    error = set_dscp(fd, dscp);
+    error = set_dscp(fd, ss.ss_family, dscp);
     if (error) {
-        VLOG_ERR("%s: socket: %s", target, sock_strerror(error));
+        VLOG_ERR("%s: set_dscp: %s", target, sock_strerror(error));
         goto error;
     }
 
@@ -925,28 +942,6 @@ error:
     return -error;
 }
 
-/* Returns a readable and writable fd for /dev/null, if successful, otherwise
- * a negative errno value.  The caller must not close the returned fd (because
- * the same fd will be handed out to subsequent callers). */
-int
-get_null_fd(void)
-{
-    static struct ovsthread_once once = OVSTHREAD_ONCE_INITIALIZER;
-    static int null_fd;
-
-    if (ovsthread_once_start(&once)) {
-        null_fd = open("/dev/null", O_RDWR);
-        if (null_fd < 0) {
-            int error = errno;
-            VLOG_ERR("could not open /dev/null: %s", ovs_strerror(error));
-            null_fd = -error;
-        }
-        ovsthread_once_done(&once);
-    }
-
-    return null_fd;
-}
-
 int
 read_fully(int fd, void *p_, size_t size, size_t *bytes_read)
 {
@@ -1050,6 +1045,7 @@ get_mtime(const char *file_name, struct timespec *mtime)
     }
 }
 
+#ifndef _WIN32
 void
 xpipe(int fds[2])
 {
@@ -1065,6 +1061,7 @@ xpipe_nonblocking(int fds[2])
     xset_nonblocking(fds[0]);
     xset_nonblocking(fds[1]);
 }
+#endif
 
 static int
 getsockopt_int(int fd, int level, int option, const char *optname, int *valuep)
@@ -1147,7 +1144,7 @@ describe_sockaddr(struct ds *string, int fd,
             }
         }
 #endif
-#if AF_PACKET && LINUX_DATAPATH
+#if __linux__
         else if (ss.ss_family == AF_PACKET) {
             struct sockaddr_ll sll;
 
@@ -1177,7 +1174,7 @@ describe_sockaddr(struct ds *string, int fd,
 }
 
 
-#ifdef LINUX_DATAPATH
+#ifdef __linux__
 static void
 put_fd_filename(struct ds *string, int fd)
 {
@@ -1222,7 +1219,7 @@ describe_fd(int fd)
                               : S_ISFIFO(s.st_mode) ? "FIFO"
                               : S_ISLNK(s.st_mode) ? "symbolic link"
                               : "unknown"));
-#ifdef LINUX_DATAPATH
+#ifdef __linux__
         put_fd_filename(&string, fd);
 #endif
     }